Merge branch 'master' into cc/summarization_patch

# Conflicts:
#	libs/langchain_v1/langchain/agents/middleware/summarization.py

.github/ISSUE_TEMPLATE/bug-report.yml

@@ -8,16 +8,15 @@ body:
       value: |
         Thank you for taking the time to file a bug report.
 
-        Use this to report BUGS in LangChain. For usage questions, feature requests and general design questions, please use the [LangChain Forum](https://forum.langchain.com/).
+        For usage questions, feature requests and general design questions, please use the [LangChain Forum](https://forum.langchain.com/).
 
-        Relevant links to check before filing a bug report to see if your issue has already been reported, fixed or
-        if there's another way to solve your problem:
+        Check these before submitting to see if your issue has already been reported, fixed or if there's another way to solve your problem:
 
-        * [LangChain Forum](https://forum.langchain.com/),
-        * [LangChain documentation with the integrated search](https://docs.langchain.com/oss/python/langchain/overview),
-        * [API Reference](https://reference.langchain.com/python/),
+        * [Documentation](https://docs.langchain.com/oss/python/langchain/overview),
+        * [API Reference Documentation](https://reference.langchain.com/python/),
         * [LangChain ChatBot](https://chat.langchain.com/)
         * [GitHub search](https://github.com/langchain-ai/langchain),
+        * [LangChain Forum](https://forum.langchain.com/),
   - type: checkboxes
     id: checks
     attributes:
@@ -36,16 +35,48 @@ body:
           required: true
         - label: This is not related to the langchain-community package.
           required: true
-        - label: I read what a minimal reproducible example is (https://stackoverflow.com/help/minimal-reproducible-example).
-          required: true
         - label: I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.
           required: true
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Which `langchain` package(s) is this bug related to? Select at least one.
+
+        Note that if the package you are reporting for is not listed here, it is not in this repository (e.g. `langchain-google-genai` is in [`langchain-ai/langchain-google`](https://github.com/langchain-ai/langchain-google/)).
+
+        Please report issues for other packages to their respective repositories.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general
   - type: textarea
     id: reproduction
     validations:
       required: true
     attributes:
-      label: Example Code
+      label: Example Code (Python)
       description: |
         Please add a self-contained, [minimal, reproducible, example](https://stackoverflow.com/help/minimal-reproducible-example) with your use case.
 
@@ -53,15 +84,12 @@ body:
 
         **Important!**
 
-        * Avoid screenshots when possible, as they are hard to read and (more importantly) don't allow others to copy-and-paste your code.
-        * Reduce your code to the minimum required to reproduce the issue if possible. This makes it much easier for others to help you.
-        * Use code tags (e.g., ```python ... ```) to correctly [format your code](https://help.github.com/en/github/writing-on-github/creating-and-highlighting-code-blocks#syntax-highlighting).
-        * INCLUDE the language label (e.g. `python`) after the first three backticks to enable syntax highlighting. (e.g., ```python rather than ```).
+        * Avoid screenshots, as they are hard to read and (more importantly) don't allow others to copy-and-paste your code.
+        * Reduce your code to the minimum required to reproduce the issue if possible.
 
+        (This will be automatically formatted into code, so no need for backticks.)
+      render: python
       placeholder: |
-        The following code:
-
-        ```python
         from langchain_core.runnables import RunnableLambda
 
         def bad_code(inputs) -> int:
@@ -69,17 +97,14 @@ body:
 
           chain = RunnableLambda(bad_code)
           chain.invoke('Hello!')
-        ```
   - type: textarea
-    id: error
-    validations:
-      required: false
     attributes:
       label: Error Message and Stack Trace (if applicable)
       description: |
-        If you are reporting an error, please include the full error message and stack trace.
-      placeholder: |
-        Exception + full stack trace
+        If you are reporting an error, please copy and paste the full error message and
+        stack trace.
+        (This will be automatically formatted into code, so no need for backticks.)
+      render: shell
   - type: textarea
     id: description
     attributes:
@@ -99,9 +124,7 @@ body:
     attributes:
       label: System Info
       description: |
-        Please share your system info with us. Do NOT skip this step and please don't trim
-        the output. Most users don't include enough information here and it makes it harder
-        for us to help you.
+        Please share your system info with us.
 
         Run the following command in your terminal and paste the output here:
 
@@ -113,8 +136,6 @@ body:
         from langchain_core import sys_info
         sys_info.print_sys_info()
         ```
-
-        alternatively, put the entire output of `pip freeze` here.
       placeholder: |
         python -m langchain_core.sys_info
     validations:

.github/ISSUE_TEMPLATE/config.yml

@@ -1,9 +1,18 @@
 blank_issues_enabled: false
 version: 2.1
 contact_links:
-  - name: 📚 Documentation
-    url: https://github.com/langchain-ai/docs/issues/new?template=langchain.yml
+  - name: 📚 Documentation issue
+    url: https://github.com/langchain-ai/docs/issues/new?template=01-langchain.yml
     about: Report an issue related to the LangChain documentation
   - name: 💬 LangChain Forum
     url:  https://forum.langchain.com/
     about: General community discussions and support
+  - name: 📚 LangChain Documentation
+    url: https://docs.langchain.com/oss/python/langchain/overview
+    about: View the official LangChain documentation
+  - name: 📚 API Reference Documentation
+    url: https://reference.langchain.com/python/
+    about: View the official LangChain API reference documentation
+  - name: 💬 LangChain Forum
+    url: https://forum.langchain.com/
+    about: Ask questions and get help from the community

.github/ISSUE_TEMPLATE/feature-request.yml

@@ -13,11 +13,11 @@ body:
         Relevant links to check before filing a feature request to see if your request has already been made or
         if there's another way to achieve what you want:
 
-        * [LangChain Forum](https://forum.langchain.com/),
-        * [LangChain documentation with the integrated search](https://docs.langchain.com/oss/python/langchain/overview),
-        * [API Reference](https://reference.langchain.com/python/),
+        * [Documentation](https://docs.langchain.com/oss/python/langchain/overview),
+        * [API Reference Documentation](https://reference.langchain.com/python/),
         * [LangChain ChatBot](https://chat.langchain.com/)
         * [GitHub search](https://github.com/langchain-ai/langchain),
+        * [LangChain Forum](https://forum.langchain.com/),
   - type: checkboxes
     id: checks
     attributes:
@@ -34,6 +34,40 @@ body:
           required: true
         - label: This is not related to the langchain-community package.
           required: true
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Which `langchain` package(s) is this request related to? Select at least one.
+
+        Note that if the package you are requesting for is not listed here, it is not in this repository (e.g. `langchain-google-genai` is in `langchain-ai/langchain`).
+
+        Please submit feature requests for other packages to their respective repositories.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general
   - type: textarea
     id: feature-description
     validations:

.github/ISSUE_TEMPLATE/privileged.yml

@@ -18,3 +18,33 @@ body:
     attributes:
       label: Issue Content
       description: Add the content of the issue here.
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Please select package(s) that this issue is related to.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general

.github/ISSUE_TEMPLATE/task.yml

@@ -25,13 +25,13 @@ body:
       label: Task Description
       description: |
         Provide a clear and detailed description of the task.
-        
+
         What needs to be done? Be specific about the scope and requirements.
       placeholder: |
         This task involves...
-        
+
         The goal is to...
-        
+
         Specific requirements:
         - ...
         - ...
@@ -43,7 +43,7 @@ body:
       label: Acceptance Criteria
       description: |
         Define the criteria that must be met for this task to be considered complete.
-        
+
         What are the specific deliverables or outcomes expected?
       placeholder: |
         This task will be complete when:
@@ -58,15 +58,15 @@ body:
       label: Context and Background
       description: |
         Provide any relevant context, background information, or links to related issues/PRs.
-        
+
         Why is this task needed? What problem does it solve?
       placeholder: |
         Background:
         - ...
-        
+
         Related issues/PRs:
         - #...
-        
+
         Additional context:
         - ...
     validations:
@@ -77,15 +77,45 @@ body:
       label: Dependencies
       description: |
         List any dependencies or blockers for this task.
-        
+
         Are there other tasks, issues, or external factors that need to be completed first?
       placeholder: |
         This task depends on:
         - [ ] Issue #...
         - [ ] PR #...
         - [ ] External dependency: ...
-        
+
         Blocked by:
         - ...
     validations:
       required: false
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Please select package(s) that this task is related to.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general

.github/PULL_REQUEST_TEMPLATE.md

@@ -1,28 +1,30 @@
 (Replace this entire block of text)
 
-Thank you for contributing to LangChain! Follow these steps to mark your pull request as ready for review. **If any of these steps are not completed, your PR will not be considered for review.**
+Read the full contributing guidelines: https://docs.langchain.com/oss/python/contributing/overview
+
+Thank you for contributing to LangChain! Follow these steps to have your pull request considered as ready for review.
+
+1. PR title: Should follow the format: TYPE(SCOPE): DESCRIPTION
 
-- [ ] **PR title**: Follows the format: {TYPE}({SCOPE}): {DESCRIPTION}
   - Examples:
+    - fix(anthropic): resolve flag parsing error
     - feat(core): add multi-tenant support
-    - fix(cli): resolve flag parsing error
-    - docs(openai): update API usage examples
-  - Allowed `{TYPE}` values:
-    - feat, fix, docs, style, refactor, perf, test, build, ci, chore, revert, release
-  - Allowed `{SCOPE}` values (optional):
-    - core, cli, langchain, standard-tests, text-splitters, docs, anthropic, chroma, deepseek, exa, fireworks, groq, huggingface, mistralai, nomic, ollama, openai, perplexity, prompty, qdrant, xai, infra
-  - Once you've written the title, please delete this checklist item; do not include it in the PR.
+    - test(openai): update API usage tests
+  - Allowed TYPE and SCOPE values: https://github.com/langchain-ai/langchain/blob/master/.github/workflows/pr_lint.yml#L15-L33
 
-- [ ] **PR message**: ***Delete this entire checklist*** and replace with
-  - **Description:** a description of the change. Include a [closing keyword](https://docs.github.com/en/issues/tracking-your-work-with-issues/using-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword) if applicable to a relevant issue.
-  - **Issue:** the issue # it fixes, if applicable (e.g. Fixes #123)
-  - **Dependencies:** any dependencies required for this change
+2. PR description:
 
-- [ ] **Lint and test**: Run `make format`, `make lint` and `make test` from the root of the package(s) you've modified. **We will not consider a PR unless these three are passing in CI.** See [contribution guidelines](https://docs.langchain.com/oss/python/contributing) for more.
+  - Write 1-2 sentences summarizing the change.
+  - If this PR addresses a specific issue, please include "Fixes #ISSUE_NUMBER" in the description to automatically close the issue when the PR is merged.
+  - If there are any breaking changes, please clearly describe them.
+  - If this PR depends on another PR being merged first, please include "Depends on #PR_NUMBER" inthe description.
+
+3. Run `make format`, `make lint` and `make test` from the root of the package(s) you've modified.
+
+  - We will not consider a PR unless these three are passing in CI.
 
 Additional guidelines:
 
-- Most PRs should not touch more than one package.
-- Please do not add dependencies to `pyproject.toml` files (even optional ones) unless they are **required** for unit tests. Likewise, please do not update the `uv.lock` files unless you are adding a required dependency.
-- Changes should be backwards compatible.
-- Make sure optional dependencies are imported within a function.
+  - We ask that if you use generative AI for your contribution, you include a disclaimer.
+  - PRs should not touch more than one package unless absolutely necessary.
+  - Do not update the `uv.lock` files unless or add dependencies to `pyproject.toml` files (even optional ones) unless you have explicit permission to do so by a maintainer.

.github/actions/poetry_setup/action.yml

@@ -1,93 +0,0 @@
-# An action for setting up poetry install with caching.
-# Using a custom action since the default action does not
-# take poetry install groups into account.
-# Action code from:
-# https://github.com/actions/setup-python/issues/505#issuecomment-1273013236
-name: poetry-install-with-caching
-description: Poetry install with support for caching of dependency groups.
-
-inputs:
-  python-version:
-    description: Python version, supporting MAJOR.MINOR only
-    required: true
-
-  poetry-version:
-    description: Poetry version
-    required: true
-
-  cache-key:
-    description: Cache key to use for manual handling of caching
-    required: true
-
-  working-directory:
-    description: Directory whose poetry.lock file should be cached
-    required: true
-
-runs:
-  using: composite
-  steps:
-    - uses: actions/setup-python@v5
-      name: Setup python ${{ inputs.python-version }}
-      id: setup-python
-      with:
-        python-version: ${{ inputs.python-version }}
-
-    - uses: actions/cache@v4
-      id: cache-bin-poetry
-      name: Cache Poetry binary - Python ${{ inputs.python-version }}
-      env:
-        SEGMENT_DOWNLOAD_TIMEOUT_MIN: "1"
-      with:
-        path: |
-          /opt/pipx/venvs/poetry
-        # This step caches the poetry installation, so make sure it's keyed on the poetry version as well.
-        key: bin-poetry-${{ runner.os }}-${{ runner.arch }}-py-${{ inputs.python-version }}-${{ inputs.poetry-version }}
-
-    - name: Refresh shell hashtable and fixup softlinks
-      if: steps.cache-bin-poetry.outputs.cache-hit == 'true'
-      shell: bash
-      env:
-        POETRY_VERSION: ${{ inputs.poetry-version }}
-        PYTHON_VERSION: ${{ inputs.python-version }}
-      run: |
-        set -eux
-
-        # Refresh the shell hashtable, to ensure correct `which` output.
-        hash -r
-
-        # `actions/cache@v3` doesn't always seem able to correctly unpack softlinks.
-        # Delete and recreate the softlinks pipx expects to have.
-        rm /opt/pipx/venvs/poetry/bin/python
-        cd /opt/pipx/venvs/poetry/bin
-        ln -s "$(which "python$PYTHON_VERSION")" python
-        chmod +x python
-        cd /opt/pipx_bin/
-        ln -s /opt/pipx/venvs/poetry/bin/poetry poetry
-        chmod +x poetry
-
-        # Ensure everything got set up correctly.
-        /opt/pipx/venvs/poetry/bin/python --version
-        /opt/pipx_bin/poetry --version
-
-    - name: Install poetry
-      if: steps.cache-bin-poetry.outputs.cache-hit != 'true'
-      shell: bash
-      env:
-        POETRY_VERSION: ${{ inputs.poetry-version }}
-        PYTHON_VERSION: ${{ inputs.python-version }}
-      # Install poetry using the python version installed by setup-python step.
-      run: pipx install "poetry==$POETRY_VERSION" --python '${{ steps.setup-python.outputs.python-path }}' --verbose
-
-    - name: Restore pip and poetry cached dependencies
-      uses: actions/cache@v4
-      env:
-        SEGMENT_DOWNLOAD_TIMEOUT_MIN: "4"
-        WORKDIR: ${{ inputs.working-directory == '' && '.' || inputs.working-directory }}
-      with:
-        path: |
-          ~/.cache/pip
-          ~/.cache/pypoetry/virtualenvs
-          ~/.cache/pypoetry/cache
-          ~/.cache/pypoetry/artifacts
-          ${{ env.WORKDIR }}/.venv
-        key: py-deps-${{ runner.os }}-${{ runner.arch }}-py-${{ inputs.python-version }}-poetry-${{ inputs.poetry-version }}-${{ inputs.cache-key }}-${{ hashFiles(format('{0}/**/poetry.lock', env.WORKDIR)) }}

.github/pr-file-labeler.yml

@@ -7,13 +7,12 @@ core:
       - any-glob-to-any-file:
           - "libs/core/**/*"
 
-langchain:
+langchain-classic:
   - changed-files:
       - any-glob-to-any-file:
           - "libs/langchain/**/*"
-          - "libs/langchain_v1/**/*"
 
-v1:
+langchain:
   - changed-files:
       - any-glob-to-any-file:
           - "libs/langchain_v1/**/*"
@@ -28,6 +27,11 @@ standard-tests:
       - any-glob-to-any-file:
           - "libs/standard-tests/**/*"
 
+model-profiles:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/model-profiles/**/*"
+
 text-splitters:
   - changed-files:
       - any-glob-to-any-file:
@@ -39,6 +43,81 @@ integration:
       - any-glob-to-any-file:
           - "libs/partners/**/*"
 
+anthropic:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/anthropic/**/*"
+
+chroma:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/chroma/**/*"
+
+deepseek:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/deepseek/**/*"
+
+exa:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/exa/**/*"
+
+fireworks:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/fireworks/**/*"
+
+groq:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/groq/**/*"
+
+huggingface:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/huggingface/**/*"
+
+mistralai:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/mistralai/**/*"
+
+nomic:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/nomic/**/*"
+
+ollama:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/ollama/**/*"
+
+openai:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/openai/**/*"
+
+perplexity:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/perplexity/**/*"
+
+prompty:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/prompty/**/*"
+
+qdrant:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/qdrant/**/*"
+
+xai:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/xai/**/*"
+
 # Infrastructure and DevOps
 infra:
   - changed-files:

.github/pr-title-labeler.yml

@@ -1,41 +0,0 @@
-# PR title labeler config
-#
-# Labels PRs based on conventional commit patterns in titles
-#
-# Format: type(scope): description or type!: description (breaking)
-
-add-missing-labels: true
-clear-prexisting: false
-include-commits: false
-include-title: true
-label-for-breaking-changes: breaking
-
-label-mapping:
-  documentation: ["docs"]
-  feature: ["feat"]
-  fix: ["fix"]
-  infra: ["build", "ci", "chore"]
-  integration:
-    [
-      "anthropic",
-      "chroma",
-      "deepseek",
-      "exa",
-      "fireworks",
-      "groq",
-      "huggingface",
-      "mistralai",
-      "nomic",
-      "ollama",
-      "openai",
-      "perplexity",
-      "prompty",
-      "qdrant",
-      "xai",
-    ]
-  linting: ["style"]
-  performance: ["perf"]
-  refactor: ["refactor"]
-  release: ["release"]
-  revert: ["revert"]
-  tests: ["test"]

.github/scripts/check_diff.py

@@ -30,6 +30,7 @@ LANGCHAIN_DIRS = [
     "libs/text-splitters",
     "libs/langchain",
     "libs/langchain_v1",
+    "libs/model-profiles",
 ]
 
 # When set to True, we are ignoring core dependents
@@ -134,7 +135,7 @@ def _get_configs_for_single_dir(job: str, dir_: str) -> List[Dict[str, str]]:
     elif dir_ == "libs/core":
         py_versions = ["3.10", "3.11", "3.12", "3.13", "3.14"]
     # custom logic for specific directories
-    elif dir_ in {"libs/partners/chroma", "libs/partners/nomic"}:
+    elif dir_ in {"libs/partners/chroma"}:
         py_versions = ["3.10", "3.13"]
     else:
         py_versions = ["3.10", "3.14"]

.github/scripts/get_min_versions.py

@@ -98,7 +98,7 @@ def _check_python_version_from_requirement(
         return True
     else:
         marker_str = str(requirement.marker)
-        if "python_version" or "python_full_version" in marker_str:
+        if "python_version" in marker_str or "python_full_version" in marker_str:
             python_version_str = "".join(
                 char
                 for char in marker_str

.github/workflows/_compile_integration_test.yml

@@ -35,7 +35,7 @@ jobs:
     timeout-minutes: 20
     name: "Python ${{ inputs.python-version }}"
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/_lint.yml

@@ -38,7 +38,7 @@ jobs:
     timeout-minutes: 20
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/_release.yml

@@ -54,7 +54,7 @@ jobs:
       version: ${{ steps.check-version.outputs.version }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: Set up Python + uv
         uses: "./.github/actions/uv_setup"
@@ -77,7 +77,7 @@ jobs:
         working-directory: ${{ inputs.working-directory }}
 
       - name: Upload build
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v5
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/
@@ -105,7 +105,7 @@ jobs:
     outputs:
       release-body: ${{ steps.generate-release-body.outputs.release-body }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain
           path: langchain
@@ -149,8 +149,8 @@ jobs:
             fi
           fi
 
-          # if PREV_TAG is empty, let it be empty
-          if [ -z "$PREV_TAG" ]; then
+          # if PREV_TAG is empty or came out to 0.0.0, let it be empty
+          if [ -z "$PREV_TAG" ] || [ "$PREV_TAG" = "$PKG_NAME==0.0.0" ]; then
             echo "No previous tag found - first release"
           else
             # confirm prev-tag actually exists in git repo with git tag
@@ -179,8 +179,8 @@ jobs:
           PREV_TAG: ${{ steps.check-tags.outputs.prev-tag }}
         run: |
           PREAMBLE="Changes since $PREV_TAG"
-          # if PREV_TAG is empty, then we are releasing the first version
-          if [ -z "$PREV_TAG" ]; then
+          # if PREV_TAG is empty or 0.0.0, then we are releasing the first version
+          if [ -z "$PREV_TAG" ] || [ "$PREV_TAG" = "$PKG_NAME==0.0.0" ]; then
             PREAMBLE="Initial release"
             PREV_TAG=$(git rev-list --max-parents=0 HEAD)
           fi
@@ -206,9 +206,9 @@ jobs:
       id-token: write
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
-      - uses: actions/download-artifact@v5
+      - uses: actions/download-artifact@v6
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/
@@ -237,7 +237,7 @@ jobs:
       contents: read
     timeout-minutes: 20
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       # We explicitly *don't* set up caching here. This ensures our tests are
       # maximally sensitive to catching breakage.
@@ -258,7 +258,7 @@ jobs:
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v5
+      - uses: actions/download-artifact@v6
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/
@@ -377,6 +377,7 @@ jobs:
           XAI_API_KEY: ${{ secrets.XAI_API_KEY }}
           DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}
           PPLX_API_KEY: ${{ secrets.PPLX_API_KEY }}
+          LANGCHAIN_TESTS_USER_AGENT: ${{ secrets.LANGCHAIN_TESTS_USER_AGENT }}
         run: make integration_tests
         working-directory: ${{ inputs.working-directory }}
 
@@ -395,7 +396,7 @@ jobs:
       contents: read
     strategy:
       matrix:
-        partner: [openai, anthropic]
+        partner: [anthropic]
       fail-fast: false # Continue testing other partners if one fails
     env:
       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
@@ -409,8 +410,9 @@ jobs:
       AZURE_OPENAI_LEGACY_CHAT_DEPLOYMENT_NAME: ${{ secrets.AZURE_OPENAI_LEGACY_CHAT_DEPLOYMENT_NAME }}
       AZURE_OPENAI_LLM_DEPLOYMENT_NAME: ${{ secrets.AZURE_OPENAI_LLM_DEPLOYMENT_NAME }}
       AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT_NAME: ${{ secrets.AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT_NAME }}
+      LANGCHAIN_TESTS_USER_AGENT: ${{ secrets.LANGCHAIN_TESTS_USER_AGENT }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       # We implement this conditional as Github Actions does not have good support
       # for conditionally needing steps. https://github.com/actions/runner/issues/491
@@ -428,7 +430,7 @@ jobs:
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v5
+      - uses: actions/download-artifact@v6
         if: startsWith(inputs.working-directory, 'libs/core')
         with:
           name: dist
@@ -442,7 +444,7 @@ jobs:
             git ls-remote --tags origin "langchain-${{ matrix.partner }}*" \
             | awk '{print $2}' \
             | sed 's|refs/tags/||' \
-            | grep -E '[0-9]+\.[0-9]+\.[0-9]+([a-zA-Z]+[0-9]+)?$' \
+            | grep -E '[0-9]+\.[0-9]+\.[0-9]+$' \
             | sort -Vr \
             | head -n 1
           )"
@@ -490,14 +492,14 @@ jobs:
         working-directory: ${{ inputs.working-directory }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: Set up Python + uv
         uses: "./.github/actions/uv_setup"
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v5
+      - uses: actions/download-artifact@v6
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/
@@ -530,14 +532,14 @@ jobs:
         working-directory: ${{ inputs.working-directory }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: Set up Python + uv
         uses: "./.github/actions/uv_setup"
         with:
           python-version: ${{ env.PYTHON_VERSION }}
 
-      - uses: actions/download-artifact@v5
+      - uses: actions/download-artifact@v6
         with:
           name: dist
           path: ${{ inputs.working-directory }}/dist/

.github/workflows/_test.yml

@@ -33,7 +33,7 @@ jobs:
     name: "Python ${{ inputs.python-version }}"
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/_test_pydantic.yml

@@ -36,7 +36,7 @@ jobs:
     name: "Pydantic ~=${{ inputs.pydantic-version }}"
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/auto-label-by-package.yml

@@ -0,0 +1,107 @@
+name: Auto Label Issues by Package
+
+on:
+  issues:
+    types: [opened, edited]
+
+jobs:
+  label-by-package:
+    permissions:
+      issues: write
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Sync package labels
+        uses: actions/github-script@v8
+        with:
+          script: |
+            const body = context.payload.issue.body || "";
+
+            // Extract text under "### Package"
+            const match = body.match(/### Package\s+([\s\S]*?)\n###/i);
+            if (!match) return;
+
+            const packageSection = match[1].trim();
+
+            // Mapping table for package names to labels
+            const mapping = {
+              "langchain": "langchain",
+              "langchain-openai": "openai",
+              "langchain-anthropic": "anthropic",
+              "langchain-classic": "langchain-classic",
+              "langchain-core": "core",
+              "langchain-cli": "cli",
+              "langchain-model-profiles": "model-profiles",
+              "langchain-tests": "standard-tests",
+              "langchain-text-splitters": "text-splitters",
+              "langchain-chroma": "chroma",
+              "langchain-deepseek": "deepseek",
+              "langchain-exa": "exa",
+              "langchain-fireworks": "fireworks",
+              "langchain-groq": "groq",
+              "langchain-huggingface": "huggingface",
+              "langchain-mistralai": "mistralai",
+              "langchain-nomic": "nomic",
+              "langchain-ollama": "ollama",
+              "langchain-perplexity": "perplexity",
+              "langchain-prompty": "prompty",
+              "langchain-qdrant": "qdrant",
+              "langchain-xai": "xai",
+            };
+
+            // All possible package labels we manage
+            const allPackageLabels = Object.values(mapping);
+            const selectedLabels = [];
+
+            // Check if this is checkbox format (multiple selection)
+            const checkboxMatches = packageSection.match(/- \[x\]\s+([^\n\r]+)/gi);
+            if (checkboxMatches) {
+              // Handle checkbox format
+              for (const match of checkboxMatches) {
+                const packageName = match.replace(/- \[x\]\s+/i, '').trim();
+                const label = mapping[packageName];
+                if (label && !selectedLabels.includes(label)) {
+                  selectedLabels.push(label);
+                }
+              }
+            } else {
+              // Handle dropdown format (single selection)
+              const label = mapping[packageSection];
+              if (label) {
+                selectedLabels.push(label);
+              }
+            }
+
+            // Get current issue labels
+            const issue = await github.rest.issues.get({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              issue_number: context.issue.number
+            });
+
+            const currentLabels = issue.data.labels.map(label => label.name);
+            const currentPackageLabels = currentLabels.filter(label => allPackageLabels.includes(label));
+
+            // Determine labels to add and remove
+            const labelsToAdd = selectedLabels.filter(label => !currentPackageLabels.includes(label));
+            const labelsToRemove = currentPackageLabels.filter(label => !selectedLabels.includes(label));
+
+            // Add new labels
+            if (labelsToAdd.length > 0) {
+              await github.rest.issues.addLabels({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: context.issue.number,
+                labels: labelsToAdd
+              });
+            }
+
+            // Remove old labels
+            for (const label of labelsToRemove) {
+              await github.rest.issues.removeLabel({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: context.issue.number,
+                name: label
+              });
+            }

.github/workflows/check_core_versions.yml

@@ -18,7 +18,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "✅ Verify pyproject.toml & version.py Match"
         run: |

.github/workflows/check_diffs.yml

@@ -47,7 +47,7 @@ jobs:
     if: ${{ !contains(github.event.pull_request.labels.*.name, 'ci-ignore') }}
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
       - name: "🐍 Setup Python 3.11"
         uses: actions/setup-python@v6
         with:
@@ -141,7 +141,7 @@ jobs:
       run:
         working-directory: ${{ matrix.job-configs.working-directory }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ matrix.job-configs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"
@@ -182,7 +182,7 @@ jobs:
         job-configs: ${{ fromJson(needs.build.outputs.codspeed) }}
       fail-fast: false
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "📦 Install UV Package Manager"
         uses: astral-sh/setup-uv@v7

.github/workflows/integration_tests.yml

@@ -71,14 +71,14 @@ jobs:
         working-directory: ${{ fromJSON(needs.compute-matrix.outputs.matrix).working-directory }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           path: langchain
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain-google
           path: langchain-google
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain-aws
           path: langchain-aws
@@ -155,6 +155,7 @@ jobs:
           WATSONX_APIKEY: ${{ secrets.WATSONX_APIKEY }}
           WATSONX_PROJECT_ID: ${{ secrets.WATSONX_PROJECT_ID }}
           XAI_API_KEY: ${{ secrets.XAI_API_KEY }}
+          LANGCHAIN_TESTS_USER_AGENT: ${{ secrets.LANGCHAIN_TESTS_USER_AGENT }}
         run: |
           cd langchain/${{ matrix.working-directory }}
           make integration_tests

.github/workflows/pr_lint.yml

@@ -26,11 +26,13 @@
 #   * revert     — reverts a previous commit
 #   * release    — prepare a new release
 #
-# Allowed Scopes (optional):
-#   core, cli, langchain, langchain_v1, langchain-classic, standard-tests,
-#   text-splitters, docs, anthropic, chroma, deepseek, exa, fireworks, groq,
-#   huggingface, mistralai, nomic, ollama, openai, perplexity, prompty, qdrant,
-#   xai, infra
+# Allowed Scope(s) (optional):
+#   core, cli, langchain, langchain_v1, langchain-classic, model-profiles,
+#   standard-tests, text-splitters, docs, anthropic, chroma, deepseek, exa,
+#   fireworks, groq, huggingface, mistralai, nomic, ollama, openai,
+#   perplexity, prompty, qdrant, xai, infra, deps
+#
+# Multiple scopes can be used by separating them with a comma.
 #
 # Rules:
 #   1. The 'Type' must start with a lowercase letter.
@@ -79,8 +81,8 @@ jobs:
             core
             cli
             langchain
-            langchain_v1
             langchain-classic
+            model-profiles
             standard-tests
             text-splitters
             docs
@@ -100,6 +102,7 @@ jobs:
             qdrant
             xai
             infra
+            deps
           requireScope: false
           disallowScopes: |
             release

.github/workflows/v03_api_doc_build.yml

@@ -23,12 +23,12 @@ jobs:
     permissions:
       contents: read
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           ref: v0.3
           path: langchain
 
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain-api-docs-html
           path: langchain-api-docs-html

.gitignore

@@ -1,6 +1,8 @@
 .vs/
 .claude/
 .idea/
+#Emacs backup
+*~
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
@@ -161,3 +163,6 @@ node_modules
 
 prof
 virtualenv/
+scratch/
+
+.langgraph_api/

.mcp.json

@@ -0,0 +1,8 @@
+{
+  "mcpServers": {
+    "docs-langchain": {
+      "type": "http",
+      "url": "https://docs.langchain.com/mcp"
+    }
+  }
+}

AGENTS.md

@@ -1,255 +1,58 @@
-# Global Development Guidelines for LangChain Projects
+# Global development guidelines for the LangChain monorepo
 
-## Core Development Principles
+This document provides context to understand the LangChain Python project and assist with development.
 
-### 1. Maintain Stable Public Interfaces ⚠️ CRITICAL
+## Project architecture and context
 
-**Always attempt to preserve function signatures, argument positions, and names for exported/public methods.**
+### Monorepo structure
 
-❌ **Bad - Breaking Change:**
+This is a Python monorepo with multiple independently versioned packages that use `uv`.
 
-```python
-def get_user(id, verbose=False):  # Changed from `user_id`
-    pass
+```txt
+langchain/
+├── libs/
+│   ├── core/             # `langchain-core` primitives and base abstractions
+│   ├── langchain/        # `langchain-classic` (legacy, no new features)
+│   ├── langchain_v1/     # Actively maintained `langchain` package
+│   ├── partners/         # Third-party integrations
+│   │   ├── openai/       # OpenAI models and embeddings
+│   │   ├── anthropic/    # Anthropic (Claude) integration
+│   │   ├── ollama/       # Local model support
+│   │   └── ... (other integrations maintained by the LangChain team)
+│   ├── text-splitters/   # Document chunking utilities
+│   ├── standard-tests/   # Shared test suite for integrations
+│   ├── model-profiles/   # Model configuration profiles
+│   └── cli/              # Command-line interface tools
+├── .github/              # CI/CD workflows and templates
+├── .vscode/              # VSCode IDE standard settings and recommended extensions
+└── README.md             # Information about LangChain
 ```
 
-✅ **Good - Stable Interface:**
+- **Core layer** (`langchain-core`): Base abstractions, interfaces, and protocols. Users should not need to know about this layer directly.
+- **Implementation layer** (`langchain`): Concrete implementations and high-level public utilities
+- **Integration layer** (`partners/`): Third-party service integrations. Note that this monorepo is not exhaustive of all LangChain integrations; some are maintained in separate repos, such as `langchain-ai/langchain-google` and `langchain-ai/langchain-aws`. Usually these repos are cloned at the same level as this monorepo, so if needed, you can refer to their code directly by navigating to `../langchain-google/` from this monorepo.
+- **Testing layer** (`standard-tests/`): Standardized integration tests for partner integrations
 
-```python
-def get_user(user_id: str, verbose: bool = False) -> User:
-    """Retrieve user by ID with optional verbose output."""
-    pass
-```
+### Development tools & commands**
 
-**Before making ANY changes to public APIs:**
+- `uv` – Fast Python package installer and resolver (replaces pip/poetry)
+- `make` – Task runner for common development commands. Feel free to look at the `Makefile` for available commands and usage patterns.
+- `ruff` – Fast Python linter and formatter
+- `mypy` – Static type checking
+- `pytest` – Testing framework
 
-- Check if the function/class is exported in `__init__.py`
-- Look for existing usage patterns in tests and examples
-- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
-- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+This monorepo uses `uv` for dependency management. Local development uses editable installs: `[tool.uv.sources]`
 
-🧠 *Ask yourself:* "Would this change break someone's code if they used it last week?"
-
-### 2. Code Quality Standards
-
-**All Python code MUST include type hints and return types.**
-
-❌ **Bad:**
-
-```python
-def p(u, d):
-    return [x for x in u if x not in d]
-```
-
-✅ **Good:**
-
-```python
-def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
-    """Filter out users that are not in the known users set.
-
-    Args:
-        users: List of user identifiers to filter.
-        known_users: Set of known/valid user identifiers.
-
-    Returns:
-        List of users that are not in the known_users set.
-    """
-    return [user for user in users if user not in known_users]
-```
-
-**Style Requirements:**
-
-- Use descriptive, **self-explanatory variable names**. Avoid overly short or cryptic identifiers.
-- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
-- Avoid unnecessary abstraction or premature optimization
-- Follow existing patterns in the codebase you're modifying
-
-### 3. Testing Requirements
-
-**Every new feature or bugfix MUST be covered by unit tests.**
-
-**Test Organization:**
-
-- Unit tests: `tests/unit_tests/` (no network calls allowed)
-- Integration tests: `tests/integration_tests/` (network calls permitted)
-- Use `pytest` as the testing framework
-
-**Test Quality Checklist:**
-
-- [ ] Tests fail when your new logic is broken
-- [ ] Happy path is covered
-- [ ] Edge cases and error conditions are tested
-- [ ] Use fixtures/mocks for external dependencies
-- [ ] Tests are deterministic (no flaky tests)
-
-Checklist questions:
-
-- [ ] Does the test suite fail if your new logic is broken?
-- [ ] Are all expected behaviors exercised (happy path, invalid input, etc)?
-- [ ] Do tests use fixtures or mocks where needed?
-
-```python
-def test_filter_unknown_users():
-    """Test filtering unknown users from a list."""
-    users = ["alice", "bob", "charlie"]
-    known_users = {"alice", "bob"}
-
-    result = filter_unknown_users(users, known_users)
-
-    assert result == ["charlie"]
-    assert len(result) == 1
-```
-
-### 4. Security and Risk Assessment
-
-**Security Checklist:**
-
-- No `eval()`, `exec()`, or `pickle` on user-controlled input
-- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
-- Remove unreachable/commented code before committing
-- Race conditions or resource leaks (file handles, sockets, threads).
-- Ensure proper resource cleanup (file handles, connections)
-
-❌ **Bad:**
-
-```python
-def load_config(path):
-    with open(path) as f:
-        return eval(f.read())  # ⚠️ Never eval config
-```
-
-✅ **Good:**
-
-```python
-import json
-
-def load_config(path: str) -> dict:
-    with open(path) as f:
-        return json.load(f)
-```
-
-### 5. Documentation Standards
-
-**Use Google-style docstrings with Args section for all public functions.**
-
-❌ **Insufficient Documentation:**
-
-```python
-def send_email(to, msg):
-    """Send an email to a recipient."""
-```
-
-✅ **Complete Documentation:**
-
-```python
-def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
-    """
-    Send an email to a recipient with specified priority.
-
-    Args:
-        to: The email address of the recipient.
-        msg: The message body to send.
-        priority: Email priority level (`'low'`, `'normal'`, `'high'`).
-
-    Returns:
-        `True` if email was sent successfully, `False` otherwise.
-
-    Raises:
-        `InvalidEmailError`: If the email address format is invalid.
-        `SMTPConnectionError`: If unable to connect to email server.
-    """
-```
-
-**Documentation Guidelines:**
-
-- Types go in function signatures, NOT in docstrings
-  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
-- Focus on "why" rather than "what" in descriptions
-- Document all parameters, return values, and exceptions
-- Keep descriptions concise but clear
-- Ensure American English spelling (e.g., "behavior", not "behaviour")
-
-📌 *Tip:* Keep descriptions concise but clear. Only document return values if non-obvious.
-
-### 6. Architectural Improvements
-
-**When you encounter code that could be improved, suggest better designs:**
-
-❌ **Poor Design:**
-
-```python
-def process_data(data, db_conn, email_client, logger):
-    # Function doing too many things
-    validated = validate_data(data)
-    result = db_conn.save(validated)
-    email_client.send_notification(result)
-    logger.log(f"Processed {len(data)} items")
-    return result
-```
-
-✅ **Better Design:**
-
-```python
-@dataclass
-class ProcessingResult:
-    """Result of data processing operation."""
-    items_processed: int
-    success: bool
-    errors: List[str] = field(default_factory=list)
-
-class DataProcessor:
-    """Handles data validation, storage, and notification."""
-
-    def __init__(self, db_conn: Database, email_client: EmailClient):
-        self.db = db_conn
-        self.email = email_client
-
-    def process(self, data: List[dict]) -> ProcessingResult:
-        """Process and store data with notifications."""
-        validated = self._validate_data(data)
-        result = self.db.save(validated)
-        self._notify_completion(result)
-        return result
-```
-
-**Design Improvement Areas:**
-
-If there's a **cleaner**, **more scalable**, or **simpler** design, highlight it and suggest improvements that would:
-
-- Reduce code duplication through shared utilities
-- Make unit testing easier
-- Improve separation of concerns (single responsibility)
-- Make unit testing easier through dependency injection
-- Add clarity without adding complexity
-- Prefer dataclasses for structured data
-
-## Development Tools & Commands
-
-### Package Management
-
-```bash
-# Add package
-uv add package-name
-
-# Sync project dependencies
-uv sync
-uv lock
-```
-
-### Testing
+Each package in `libs/` has its own `pyproject.toml` and `uv.lock`.
 
 ```bash
 # Run unit tests (no network)
 make test
 
-# Don't run integration tests, as API keys must be set
-
 # Run specific test file
 uv run --group test pytest tests/unit_tests/test_specific.py
 ```
 
-### Code Quality
-
 ```bash
 # Lint code
 make lint
@@ -261,66 +64,118 @@ make format
 uv run --group lint mypy .
 ```
 
-### Dependency Management Patterns
+#### Key config files
 
-**Local Development Dependencies:**
+- pyproject.toml: Main workspace configuration with dependency groups
+- uv.lock: Locked dependencies for reproducible builds
+- Makefile: Development tasks
 
-```toml
-[tool.uv.sources]
-langchain-core = { path = "../core", editable = true }
-langchain-tests = { path = "../standard-tests", editable = true }
-```
+#### Commit standards
 
-**For tools, use the `@tool` decorator from `langchain_core.tools`:**
+Suggest PR titles that follow Conventional Commits format. Refer to .github/workflows/pr_lint for allowed types and scopes.
 
-```python
-from langchain_core.tools import tool
+#### Pull request guidelines
 
-@tool
-def search_database(query: str) -> str:
-    """Search the database for relevant information.
+- Always add a disclaimer to the PR description mentioning how AI agents are involved with the contribution.
+- Describe the "why" of the changes, why the proposed solution is the right one. Limit prose.
+- Highlight areas of the proposed changes that require careful review.
+
+## Core development principles
+
+### Maintain stable public interfaces
+
+CRITICAL: Always attempt to preserve function signatures, argument positions, and names for exported/public methods. Do not make breaking changes.
+
+**Before making ANY changes to public APIs:**
+
+- Check if the function/class is exported in `__init__.py`
+- Look for existing usage patterns in tests and examples
+- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
+- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+
+Ask: "Would this change break someone's code if they used it last week?"
+
+### Code quality standards
+
+All Python code MUST include type hints and return types.
+
+```python title="Example"
+def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
+    """Single line description of the function.
+
+    Any additional context about the function can go here.
 
     Args:
-        query: The search query string.
+        users: List of user identifiers to filter.
+        known_users: Set of known/valid user identifiers.
+
+    Returns:
+        List of users that are not in the known_users set.
     """
-    # Implementation here
-    return results
 ```
 
-## Commit Standards
+- Use descriptive, self-explanatory variable names.
+- Follow existing patterns in the codebase you're modifying
+- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
 
-**Use Conventional Commits format for PR titles:**
+### Testing requirements
 
-- `feat(core): add multi-tenant support`
-- `fix(cli): resolve flag parsing error`
-- `docs: update API usage examples`
-- `docs(openai): update API usage examples`
+Every new feature or bugfix MUST be covered by unit tests.
 
-## Framework-Specific Guidelines
+- Unit tests: `tests/unit_tests/` (no network calls allowed)
+- Integration tests: `tests/integration_tests/` (network calls permitted)
+- We use `pytest` as the testing framework; if in doubt, check other existing tests for examples.
+- The testing file structure should mirror the source code structure.
 
-- Follow the existing patterns in `langchain-core` for base abstractions
-- Use `langchain_core.callbacks` for execution tracking
-- Implement proper streaming support where applicable
-- Avoid deprecated components like legacy `LLMChain`
+**Checklist:**
 
-### Partner Integrations
+- [ ] Tests fail when your new logic is broken
+- [ ] Happy path is covered
+- [ ] Edge cases and error conditions are tested
+- [ ] Use fixtures/mocks for external dependencies
+- [ ] Tests are deterministic (no flaky tests)
+- [ ] Does the test suite fail if your new logic is broken?
 
-- Follow the established patterns in existing partner libraries
-- Implement standard interfaces (`BaseChatModel`, `BaseEmbeddings`, etc.)
-- Include comprehensive integration tests
-- Document API key requirements and authentication
+### Security and risk assessment
 
----
+- No `eval()`, `exec()`, or `pickle` on user-controlled input
+- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
+- Remove unreachable/commented code before committing
+- Race conditions or resource leaks (file handles, sockets, threads).
+- Ensure proper resource cleanup (file handles, connections)
 
-## Quick Reference Checklist
+### Documentation standards
 
-Before submitting code changes:
+Use Google-style docstrings with Args section for all public functions.
 
-- [ ] **Breaking Changes**: Verified no public API changes
-- [ ] **Type Hints**: All functions have complete type annotations
-- [ ] **Tests**: New functionality is fully tested
-- [ ] **Security**: No dangerous patterns (eval, silent failures, etc.)
-- [ ] **Documentation**: Google-style docstrings for public functions
-- [ ] **Code Quality**: `make lint` and `make format` pass
-- [ ] **Architecture**: Suggested improvements where applicable
-- [ ] **Commit Message**: Follows Conventional Commits format
+```python title="Example"
+def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
+    """Send an email to a recipient with specified priority.
+
+    Any additional context about the function can go here.
+
+    Args:
+        to: The email address of the recipient.
+        msg: The message body to send.
+        priority: Email priority level.
+
+    Returns:
+        `True` if email was sent successfully, `False` otherwise.
+
+    Raises:
+        InvalidEmailError: If the email address format is invalid.
+        SMTPConnectionError: If unable to connect to email server.
+    """
+```
+
+- Types go in function signatures, NOT in docstrings
+  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
+- Focus on "why" rather than "what" in descriptions
+- Document all parameters, return values, and exceptions
+- Keep descriptions concise but clear
+- Ensure American English spelling (e.g., "behavior", not "behaviour")
+
+## Additional resources
+
+- **Documentation:** https://docs.langchain.com/oss/python/langchain/overview and source at https://github.com/langchain-ai/docs or `../docs/`. Prefer the local install and use file search tools for best results. If needed, use the docs MCP server as defined in `.mcp.json` for programmatic access.
+- **Contributing Guide:** [`.github/CONTRIBUTING.md`](https://docs.langchain.com/oss/python/contributing/overview)

CLAUDE.md

@@ -1,255 +1,58 @@
-# Global Development Guidelines for LangChain Projects
+# Global development guidelines for the LangChain monorepo
 
-## Core Development Principles
+This document provides context to understand the LangChain Python project and assist with development.
 
-### 1. Maintain Stable Public Interfaces ⚠️ CRITICAL
+## Project architecture and context
 
-**Always attempt to preserve function signatures, argument positions, and names for exported/public methods.**
+### Monorepo structure
 
-❌ **Bad - Breaking Change:**
+This is a Python monorepo with multiple independently versioned packages that use `uv`.
 
-```python
-def get_user(id, verbose=False):  # Changed from `user_id`
-    pass
+```txt
+langchain/
+├── libs/
+│   ├── core/             # `langchain-core` primitives and base abstractions
+│   ├── langchain/        # `langchain-classic` (legacy, no new features)
+│   ├── langchain_v1/     # Actively maintained `langchain` package
+│   ├── partners/         # Third-party integrations
+│   │   ├── openai/       # OpenAI models and embeddings
+│   │   ├── anthropic/    # Anthropic (Claude) integration
+│   │   ├── ollama/       # Local model support
+│   │   └── ... (other integrations maintained by the LangChain team)
+│   ├── text-splitters/   # Document chunking utilities
+│   ├── standard-tests/   # Shared test suite for integrations
+│   ├── model-profiles/   # Model configuration profiles
+│   └── cli/              # Command-line interface tools
+├── .github/              # CI/CD workflows and templates
+├── .vscode/              # VSCode IDE standard settings and recommended extensions
+└── README.md             # Information about LangChain
 ```
 
-✅ **Good - Stable Interface:**
+- **Core layer** (`langchain-core`): Base abstractions, interfaces, and protocols. Users should not need to know about this layer directly.
+- **Implementation layer** (`langchain`): Concrete implementations and high-level public utilities
+- **Integration layer** (`partners/`): Third-party service integrations. Note that this monorepo is not exhaustive of all LangChain integrations; some are maintained in separate repos, such as `langchain-ai/langchain-google` and `langchain-ai/langchain-aws`. Usually these repos are cloned at the same level as this monorepo, so if needed, you can refer to their code directly by navigating to `../langchain-google/` from this monorepo.
+- **Testing layer** (`standard-tests/`): Standardized integration tests for partner integrations
 
-```python
-def get_user(user_id: str, verbose: bool = False) -> User:
-    """Retrieve user by ID with optional verbose output."""
-    pass
-```
+### Development tools & commands**
 
-**Before making ANY changes to public APIs:**
+- `uv` – Fast Python package installer and resolver (replaces pip/poetry)
+- `make` – Task runner for common development commands. Feel free to look at the `Makefile` for available commands and usage patterns.
+- `ruff` – Fast Python linter and formatter
+- `mypy` – Static type checking
+- `pytest` – Testing framework
 
-- Check if the function/class is exported in `__init__.py`
-- Look for existing usage patterns in tests and examples
-- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
-- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+This monorepo uses `uv` for dependency management. Local development uses editable installs: `[tool.uv.sources]`
 
-🧠 *Ask yourself:* "Would this change break someone's code if they used it last week?"
-
-### 2. Code Quality Standards
-
-**All Python code MUST include type hints and return types.**
-
-❌ **Bad:**
-
-```python
-def p(u, d):
-    return [x for x in u if x not in d]
-```
-
-✅ **Good:**
-
-```python
-def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
-    """Filter out users that are not in the known users set.
-
-    Args:
-        users: List of user identifiers to filter.
-        known_users: Set of known/valid user identifiers.
-
-    Returns:
-        List of users that are not in the known_users set.
-    """
-    return [user for user in users if user not in known_users]
-```
-
-**Style Requirements:**
-
-- Use descriptive, **self-explanatory variable names**. Avoid overly short or cryptic identifiers.
-- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
-- Avoid unnecessary abstraction or premature optimization
-- Follow existing patterns in the codebase you're modifying
-
-### 3. Testing Requirements
-
-**Every new feature or bugfix MUST be covered by unit tests.**
-
-**Test Organization:**
-
-- Unit tests: `tests/unit_tests/` (no network calls allowed)
-- Integration tests: `tests/integration_tests/` (network calls permitted)
-- Use `pytest` as the testing framework
-
-**Test Quality Checklist:**
-
-- [ ] Tests fail when your new logic is broken
-- [ ] Happy path is covered
-- [ ] Edge cases and error conditions are tested
-- [ ] Use fixtures/mocks for external dependencies
-- [ ] Tests are deterministic (no flaky tests)
-
-Checklist questions:
-
-- [ ] Does the test suite fail if your new logic is broken?
-- [ ] Are all expected behaviors exercised (happy path, invalid input, etc)?
-- [ ] Do tests use fixtures or mocks where needed?
-
-```python
-def test_filter_unknown_users():
-    """Test filtering unknown users from a list."""
-    users = ["alice", "bob", "charlie"]
-    known_users = {"alice", "bob"}
-
-    result = filter_unknown_users(users, known_users)
-
-    assert result == ["charlie"]
-    assert len(result) == 1
-```
-
-### 4. Security and Risk Assessment
-
-**Security Checklist:**
-
-- No `eval()`, `exec()`, or `pickle` on user-controlled input
-- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
-- Remove unreachable/commented code before committing
-- Race conditions or resource leaks (file handles, sockets, threads).
-- Ensure proper resource cleanup (file handles, connections)
-
-❌ **Bad:**
-
-```python
-def load_config(path):
-    with open(path) as f:
-        return eval(f.read())  # ⚠️ Never eval config
-```
-
-✅ **Good:**
-
-```python
-import json
-
-def load_config(path: str) -> dict:
-    with open(path) as f:
-        return json.load(f)
-```
-
-### 5. Documentation Standards
-
-**Use Google-style docstrings with Args section for all public functions.**
-
-❌ **Insufficient Documentation:**
-
-```python
-def send_email(to, msg):
-    """Send an email to a recipient."""
-```
-
-✅ **Complete Documentation:**
-
-```python
-def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
-    """
-    Send an email to a recipient with specified priority.
-
-    Args:
-        to: The email address of the recipient.
-        msg: The message body to send.
-        priority: Email priority level (`'low'`, `'normal'`, `'high'`).
-
-    Returns:
-        `True` if email was sent successfully, `False` otherwise.
-
-    Raises:
-        `InvalidEmailError`: If the email address format is invalid.
-        `SMTPConnectionError`: If unable to connect to email server.
-    """
-```
-
-**Documentation Guidelines:**
-
-- Types go in function signatures, NOT in docstrings
-  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
-- Focus on "why" rather than "what" in descriptions
-- Document all parameters, return values, and exceptions
-- Keep descriptions concise but clear
-- Ensure American English spelling (e.g., "behavior", not "behaviour")
-
-📌 *Tip:* Keep descriptions concise but clear. Only document return values if non-obvious.
-
-### 6. Architectural Improvements
-
-**When you encounter code that could be improved, suggest better designs:**
-
-❌ **Poor Design:**
-
-```python
-def process_data(data, db_conn, email_client, logger):
-    # Function doing too many things
-    validated = validate_data(data)
-    result = db_conn.save(validated)
-    email_client.send_notification(result)
-    logger.log(f"Processed {len(data)} items")
-    return result
-```
-
-✅ **Better Design:**
-
-```python
-@dataclass
-class ProcessingResult:
-    """Result of data processing operation."""
-    items_processed: int
-    success: bool
-    errors: List[str] = field(default_factory=list)
-
-class DataProcessor:
-    """Handles data validation, storage, and notification."""
-
-    def __init__(self, db_conn: Database, email_client: EmailClient):
-        self.db = db_conn
-        self.email = email_client
-
-    def process(self, data: List[dict]) -> ProcessingResult:
-        """Process and store data with notifications."""
-        validated = self._validate_data(data)
-        result = self.db.save(validated)
-        self._notify_completion(result)
-        return result
-```
-
-**Design Improvement Areas:**
-
-If there's a **cleaner**, **more scalable**, or **simpler** design, highlight it and suggest improvements that would:
-
-- Reduce code duplication through shared utilities
-- Make unit testing easier
-- Improve separation of concerns (single responsibility)
-- Make unit testing easier through dependency injection
-- Add clarity without adding complexity
-- Prefer dataclasses for structured data
-
-## Development Tools & Commands
-
-### Package Management
-
-```bash
-# Add package
-uv add package-name
-
-# Sync project dependencies
-uv sync
-uv lock
-```
-
-### Testing
+Each package in `libs/` has its own `pyproject.toml` and `uv.lock`.
 
 ```bash
 # Run unit tests (no network)
 make test
 
-# Don't run integration tests, as API keys must be set
-
 # Run specific test file
 uv run --group test pytest tests/unit_tests/test_specific.py
 ```
 
-### Code Quality
-
 ```bash
 # Lint code
 make lint
@@ -261,66 +64,118 @@ make format
 uv run --group lint mypy .
 ```
 
-### Dependency Management Patterns
+#### Key config files
 
-**Local Development Dependencies:**
+- pyproject.toml: Main workspace configuration with dependency groups
+- uv.lock: Locked dependencies for reproducible builds
+- Makefile: Development tasks
 
-```toml
-[tool.uv.sources]
-langchain-core = { path = "../core", editable = true }
-langchain-tests = { path = "../standard-tests", editable = true }
-```
+#### Commit standards
 
-**For tools, use the `@tool` decorator from `langchain_core.tools`:**
+Suggest PR titles that follow Conventional Commits format. Refer to .github/workflows/pr_lint for allowed types and scopes.
 
-```python
-from langchain_core.tools import tool
+#### Pull request guidelines
 
-@tool
-def search_database(query: str) -> str:
-    """Search the database for relevant information.
+- Always add a disclaimer to the PR description mentioning how AI agents are involved with the contribution.
+- Describe the "why" of the changes, why the proposed solution is the right one. Limit prose.
+- Highlight areas of the proposed changes that require careful review.
+
+## Core development principles
+
+### Maintain stable public interfaces
+
+CRITICAL: Always attempt to preserve function signatures, argument positions, and names for exported/public methods. Do not make breaking changes.
+
+**Before making ANY changes to public APIs:**
+
+- Check if the function/class is exported in `__init__.py`
+- Look for existing usage patterns in tests and examples
+- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
+- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+
+Ask: "Would this change break someone's code if they used it last week?"
+
+### Code quality standards
+
+All Python code MUST include type hints and return types.
+
+```python title="Example"
+def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
+    """Single line description of the function.
+
+    Any additional context about the function can go here.
 
     Args:
-        query: The search query string.
+        users: List of user identifiers to filter.
+        known_users: Set of known/valid user identifiers.
+
+    Returns:
+        List of users that are not in the known_users set.
     """
-    # Implementation here
-    return results
 ```
 
-## Commit Standards
+- Use descriptive, self-explanatory variable names.
+- Follow existing patterns in the codebase you're modifying
+- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
 
-**Use Conventional Commits format for PR titles:**
+### Testing requirements
 
-- `feat(core): add multi-tenant support`
-- `fix(cli): resolve flag parsing error`
-- `docs: update API usage examples`
-- `docs(openai): update API usage examples`
+Every new feature or bugfix MUST be covered by unit tests.
 
-## Framework-Specific Guidelines
+- Unit tests: `tests/unit_tests/` (no network calls allowed)
+- Integration tests: `tests/integration_tests/` (network calls permitted)
+- We use `pytest` as the testing framework; if in doubt, check other existing tests for examples.
+- The testing file structure should mirror the source code structure.
 
-- Follow the existing patterns in `langchain-core` for base abstractions
-- Use `langchain_core.callbacks` for execution tracking
-- Implement proper streaming support where applicable
-- Avoid deprecated components like legacy `LLMChain`
+**Checklist:**
 
-### Partner Integrations
+- [ ] Tests fail when your new logic is broken
+- [ ] Happy path is covered
+- [ ] Edge cases and error conditions are tested
+- [ ] Use fixtures/mocks for external dependencies
+- [ ] Tests are deterministic (no flaky tests)
+- [ ] Does the test suite fail if your new logic is broken?
 
-- Follow the established patterns in existing partner libraries
-- Implement standard interfaces (`BaseChatModel`, `BaseEmbeddings`, etc.)
-- Include comprehensive integration tests
-- Document API key requirements and authentication
+### Security and risk assessment
 
----
+- No `eval()`, `exec()`, or `pickle` on user-controlled input
+- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
+- Remove unreachable/commented code before committing
+- Race conditions or resource leaks (file handles, sockets, threads).
+- Ensure proper resource cleanup (file handles, connections)
 
-## Quick Reference Checklist
+### Documentation standards
 
-Before submitting code changes:
+Use Google-style docstrings with Args section for all public functions.
 
-- [ ] **Breaking Changes**: Verified no public API changes
-- [ ] **Type Hints**: All functions have complete type annotations
-- [ ] **Tests**: New functionality is fully tested
-- [ ] **Security**: No dangerous patterns (eval, silent failures, etc.)
-- [ ] **Documentation**: Google-style docstrings for public functions
-- [ ] **Code Quality**: `make lint` and `make format` pass
-- [ ] **Architecture**: Suggested improvements where applicable
-- [ ] **Commit Message**: Follows Conventional Commits format
+```python title="Example"
+def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
+    """Send an email to a recipient with specified priority.
+
+    Any additional context about the function can go here.
+
+    Args:
+        to: The email address of the recipient.
+        msg: The message body to send.
+        priority: Email priority level.
+
+    Returns:
+        `True` if email was sent successfully, `False` otherwise.
+
+    Raises:
+        InvalidEmailError: If the email address format is invalid.
+        SMTPConnectionError: If unable to connect to email server.
+    """
+```
+
+- Types go in function signatures, NOT in docstrings
+  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
+- Focus on "why" rather than "what" in descriptions
+- Document all parameters, return values, and exceptions
+- Keep descriptions concise but clear
+- Ensure American English spelling (e.g., "behavior", not "behaviour")
+
+## Additional resources
+
+- **Documentation:** https://docs.langchain.com/oss/python/langchain/overview and source at https://github.com/langchain-ai/docs or `../docs/`. Prefer the local install and use file search tools for best results. If needed, use the docs MCP server as defined in `.mcp.json` for programmatic access.
+- **Contributing Guide:** [`.github/CONTRIBUTING.md`](https://docs.langchain.com/oss/python/contributing/overview)

MIGRATE.md

@@ -1,9 +0,0 @@
-# Migrating
-
-Please see the following guides for migrating LangChain code:
-
-* Migrate to [LangChain v1.0](https://docs.langchain.com/oss/python/migrate/langchain-v1)
-* Migrate to [LangChain v0.3](https://python.langchain.com/docs/versions/v0_3/)
-* Migrate to [LangChain v0.2](https://python.langchain.com/docs/versions/v0_2/)
-* Migrating from [LangChain 0.0.x Chains](https://python.langchain.com/docs/versions/migrating_chains/)
-* Upgrade to [LangGraph Memory](https://python.langchain.com/docs/versions/migrating_memory/)

README.md

@@ -1,50 +1,43 @@
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: light)" srcset=".github/images/logo-dark.svg">
-    <source media="(prefers-color-scheme: dark)" srcset=".github/images/logo-light.svg">
-    <img alt="LangChain Logo" src=".github/images/logo-dark.svg" width="80%">
-  </picture>
-</p>
+<div align="center">
+  <a href="https://www.langchain.com/">
+    <picture>
+      <source media="(prefers-color-scheme: light)" srcset=".github/images/logo-dark.svg">
+      <source media="(prefers-color-scheme: dark)" srcset=".github/images/logo-light.svg">
+      <img alt="LangChain Logo" src=".github/images/logo-dark.svg" width="80%">
+    </picture>
+  </a>
+</div>
 
-<p align="center">
-    The platform for reliable agents.
-</p>
+<div align="center">
+  <h3>The platform for reliable agents.</h3>
+</div>
 
-<p align="center">
-  <a href="https://opensource.org/licenses/MIT" target="_blank">
-      <img src="https://img.shields.io/pypi/l/langchain" alt="PyPI - License">
-  </a>
-  <a href="https://pypistats.org/packages/langchain" target="_blank">
-      <img src="https://img.shields.io/pepy/dt/langchain" alt="PyPI - Downloads">
-  </a>
-  <a href="https://pypi.org/project/langchain/#history" target="_blank">
-      <img src="https://img.shields.io/pypi/v/langchain?label=%20" alt="Version">
-  </a>
-  <a href="https://vscode.dev/redirect?url=vscode://ms-vscode-remote.remote-containers/cloneInVolume?url=https://github.com/langchain-ai/langchain" target="_blank">
-      <img src="https://img.shields.io/static/v1?label=Dev%20Containers&message=Open&color=blue&logo=visualstudiocode" alt="Open in Dev Containers">
-  </a>
-  <a href="https://codespaces.new/langchain-ai/langchain" target="_blank">
-      <img src="https://github.com/codespaces/badge.svg" alt="Open in Github Codespace" title="Open in Github Codespace" width="150" height="20">
-  </a>
-  <a href="https://codspeed.io/langchain-ai/langchain" target="_blank">
-      <img src="https://img.shields.io/endpoint?url=https://codspeed.io/badge.json" alt="CodSpeed Badge">
-  </a>
-  <a href="https://twitter.com/langchainai" target="_blank">
-      <img src="https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI" alt="Twitter / X">
-  </a>
-</p>
+<div align="center">
+  <a href="https://opensource.org/licenses/MIT" target="_blank"><img src="https://img.shields.io/pypi/l/langchain" alt="PyPI - License"></a>
+  <a href="https://pypistats.org/packages/langchain" target="_blank"><img src="https://img.shields.io/pepy/dt/langchain" alt="PyPI - Downloads"></a>
+  <a href="https://pypi.org/project/langchain/#history" target="_blank"><img src="https://img.shields.io/pypi/v/langchain?label=%20" alt="Version"></a>
+  <a href="https://vscode.dev/redirect?url=vscode://ms-vscode-remote.remote-containers/cloneInVolume?url=https://github.com/langchain-ai/langchain" target="_blank"><img src="https://img.shields.io/static/v1?label=Dev%20Containers&message=Open&color=blue&logo=visualstudiocode" alt="Open in Dev Containers"></a>
+  <a href="https://codespaces.new/langchain-ai/langchain" target="_blank"><img src="https://github.com/codespaces/badge.svg" alt="Open in Github Codespace" title="Open in Github Codespace" width="150" height="20"></a>
+  <a href="https://codspeed.io/langchain-ai/langchain" target="_blank"><img src="https://img.shields.io/endpoint?url=https://codspeed.io/badge.json" alt="CodSpeed Badge"></a>
+  <a href="https://twitter.com/langchainai" target="_blank"><img src="https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI" alt="Twitter / X"></a>
+</div>
 
-LangChain is a framework for building LLM-powered applications. It helps you chain together interoperable components and third-party integrations to simplify AI application development —  all while future-proofing decisions as the underlying technology evolves.
+LangChain is a framework for building agents and LLM-powered applications. It helps you chain together interoperable components and third-party integrations to simplify AI application development – all while future-proofing decisions as the underlying technology evolves.
 
 ```bash
 pip install langchain
 ```
 
+If you're looking for more advanced customization or agent orchestration, check out [LangGraph](https://docs.langchain.com/oss/python/langgraph/overview), our framework for building controllable agent workflows.
+
 ---
 
-**Documentation**: To learn more about LangChain, check out [the docs](https://docs.langchain.com/oss/python/langchain/overview).
+**Documentation**:
 
-If you're looking for more advanced customization or agent orchestration, check out [LangGraph](https://docs.langchain.com/oss/python/langgraph/overview), our framework for building controllable agent workflows.
+- [docs.langchain.com](https://docs.langchain.com/oss/python/langchain/overview) – Comprehensive documentation, including conceptual overviews and guides
+- [reference.langchain.com/python](https://reference.langchain.com/python) – API reference docs for LangChain packages
+
+**Discussions**: Visit the [LangChain Forum](https://forum.langchain.com) to connect with the community and share all of your technical questions, ideas, and feedback.
 
 > [!NOTE]
 > Looking for the JS/TS library? Check out [LangChain.js](https://github.com/langchain-ai/langchainjs).
@@ -55,24 +48,27 @@ LangChain helps developers build applications powered by LLMs through a standard
 
 Use LangChain for:
 
-- **Real-time data augmentation**. Easily connect LLMs to diverse data sources and external/internal systems, drawing from LangChain’s vast library of integrations with model providers, tools, vector stores, retrievers, and more.
-- **Model interoperability**. Swap models in and out as your engineering team experiments to find the best choice for your application’s needs. As the industry frontier evolves, adapt quickly — LangChain’s abstractions keep you moving without losing momentum.
+- **Real-time data augmentation**. Easily connect LLMs to diverse data sources and external/internal systems, drawing from LangChain's vast library of integrations with model providers, tools, vector stores, retrievers, and more.
+- **Model interoperability**. Swap models in and out as your engineering team experiments to find the best choice for your application's needs. As the industry frontier evolves, adapt quickly – LangChain's abstractions keep you moving without losing momentum.
+- **Rapid prototyping**. Quickly build and iterate on LLM applications with LangChain's modular, component-based architecture. Test different approaches and workflows without rebuilding from scratch, accelerating your development cycle.
+- **Production-ready features**. Deploy reliable applications with built-in support for monitoring, evaluation, and debugging through integrations like LangSmith. Scale with confidence using battle-tested patterns and best practices.
+- **Vibrant community and ecosystem**. Leverage a rich ecosystem of integrations, templates, and community-contributed components. Benefit from continuous improvements and stay up-to-date with the latest AI developments through an active open-source community.
+- **Flexible abstraction layers**. Work at the level of abstraction that suits your needs - from high-level chains for quick starts to low-level components for fine-grained control. LangChain grows with your application's complexity.
 
-## LangChain’s ecosystem
+## LangChain ecosystem
 
 While the LangChain framework can be used standalone, it also integrates seamlessly with any LangChain product, giving developers a full suite of tools when building LLM applications.
 
 To improve your LLM application development, pair LangChain with:
 
-- [LangSmith](https://www.langchain.com/langsmith) - Helpful for agent evals and observability. Debug poor-performing LLM app runs, evaluate agent trajectories, gain visibility in production, and improve performance over time.
-- [LangGraph](https://docs.langchain.com/oss/python/langgraph/overview) - Build agents that can reliably handle complex tasks with LangGraph, our low-level agent orchestration framework. LangGraph offers customizable architecture, long-term memory, and human-in-the-loop workflows — and is trusted in production by companies like LinkedIn, Uber, Klarna, and GitLab.
-- [LangGraph Platform](https://docs.langchain.com/langgraph-platform) - Deploy and scale agents effortlessly with a purpose-built deployment platform for long-running, stateful workflows. Discover, reuse, configure, and share agents across teams — and iterate quickly with visual prototyping in [LangGraph Studio](https://langchain-ai.github.io/langgraph/concepts/langgraph_studio).
+- [LangGraph](https://docs.langchain.com/oss/python/langgraph/overview) – Build agents that can reliably handle complex tasks with LangGraph, our low-level agent orchestration framework. LangGraph offers customizable architecture, long-term memory, and human-in-the-loop workflows – and is trusted in production by companies like LinkedIn, Uber, Klarna, and GitLab.
+- [Integrations](https://docs.langchain.com/oss/python/integrations/providers/overview) – List of LangChain integrations, including chat & embedding models, tools & toolkits, and more
+- [LangSmith](https://www.langchain.com/langsmith) – Helpful for agent evals and observability. Debug poor-performing LLM app runs, evaluate agent trajectories, gain visibility in production, and improve performance over time.
+- [LangSmith Deployment](https://docs.langchain.com/langsmith/deployments) – Deploy and scale agents effortlessly with a purpose-built deployment platform for long-running, stateful workflows. Discover, reuse, configure, and share agents across teams – and iterate quickly with visual prototyping in [LangSmith Studio](https://docs.langchain.com/langsmith/studio).
+- [Deep Agents](https://github.com/langchain-ai/deepagents) *(new!)* – Build agents that can plan, use subagents, and leverage file systems for complex tasks
 
 ## Additional resources
 
-- [Learn](https://docs.langchain.com/oss/python/learn): Use cases, conceptual overviews, and more.
-- [API Reference](https://reference.langchain.com/python): Detailed reference on
-navigating base packages and integrations for LangChain.
-- [Contributing Guide](https://docs.langchain.com/oss/python/contributing/overview): Learn how to contribute to LangChain and find good first issues.
-- [LangChain Forum](https://forum.langchain.com): Connect with the community and share all of your technical questions, ideas, and feedback.
-- [Chat LangChain](https://chat.langchain.com): Ask questions & chat with our documentation.
+- [API Reference](https://reference.langchain.com/python) – Detailed reference on navigating base packages and integrations for LangChain.
+- [Contributing Guide](https://docs.langchain.com/oss/python/contributing/overview) – Learn how to contribute to LangChain projects and find good first issues.
+- [Code of Conduct](https://github.com/langchain-ai/langchain/blob/master/.github/CODE_OF_CONDUCT.md) – Our community guidelines and standards for participation.

SECURITY.md

@@ -55,10 +55,10 @@ All out of scope targets defined by huntr as well as:
 * **langchain-experimental**: This repository is for experimental code and is not
   eligible for bug bounties (see [package warning](https://pypi.org/project/langchain-experimental/)), bug reports to it will be marked as interesting or waste of
   time and published with no bounty attached.
-* **tools**: Tools in either langchain or langchain-community are not eligible for bug
+* **tools**: Tools in either `langchain` or `langchain-community` are not eligible for bug
   bounties. This includes the following directories
-  * libs/langchain/langchain/tools
-  * libs/community/langchain_community/tools
+  * `libs/langchain/langchain/tools`
+  * `libs/community/langchain_community/tools`
   * Please review the [Best Practices](#best-practices)
     for more details, but generally tools interact with the real world. Developers are
     expected to understand the security implications of their code and are responsible

libs/cli/langchain_cli/integration_template/docs/vectorstores.ipynb

@@ -295,7 +295,7 @@
    "source": [
     "## TODO: Any functionality specific to this vector store\n",
     "\n",
-    "E.g. creating a persisten database to save to your disk, etc."
+    "E.g. creating a persistent database to save to your disk, etc."
    ]
   },
   {

libs/cli/langchain_cli/utils/git.py

@@ -6,9 +6,8 @@ import hashlib
 import logging
 import re
 import shutil
-from collections.abc import Sequence
 from pathlib import Path
-from typing import Any, TypedDict
+from typing import TYPE_CHECKING, Any, TypedDict
 
 from git import Repo
 
@@ -18,6 +17,9 @@ from langchain_cli.constants import (
     DEFAULT_GIT_SUBDIRECTORY,
 )
 
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
 logger = logging.getLogger(__name__)
 
 

libs/cli/tests/unit_tests/migrate/cli_runner/case.py

@@ -1,9 +1,11 @@
 from __future__ import annotations
 
 from dataclasses import dataclass
+from typing import TYPE_CHECKING
 
-from .file import File
-from .folder import Folder
+if TYPE_CHECKING:
+    from .file import File
+    from .folder import Folder
 
 
 @dataclass

libs/cli/tests/unit_tests/migrate/cli_runner/folder.py

@@ -1,9 +1,12 @@
 from __future__ import annotations
 
-from pathlib import Path
+from typing import TYPE_CHECKING
 
 from .file import File
 
+if TYPE_CHECKING:
+    from pathlib import Path
+
 
 class Folder:
     def __init__(self, name: str, *files: Folder | File) -> None:

libs/core/README.md

@@ -34,7 +34,7 @@ The LangChain ecosystem is built on top of `langchain-core`. Some of the benefit
 
 ## 📖 Documentation
 
-For full documentation, see the [API reference](https://reference.langchain.com/python/langchain_core/).
+For full documentation, see the [API reference](https://reference.langchain.com/python/langchain_core/). For conceptual guides, tutorials, and examples on using LangChain, see the [LangChain Docs](https://docs.langchain.com/oss/python/langchain/overview).
 
 ## 📕 Releases & Versioning
 

libs/core/langchain_core/agents.py

@@ -5,12 +5,10 @@
 
 !!! warning
     New agents should be built using the
-    [langgraph library](https://github.com/langchain-ai/langgraph), which provides a
+    [`langchain` library](https://pypi.org/project/langchain/), which provides a
     simpler and more flexible way to define agents.
 
-    Please see the
-    [migration guide](https://python.langchain.com/docs/how_to/migrate_agent/) for
-    information on how to migrate existing agents to modern langgraph agents.
+    See docs on [building agents](https://docs.langchain.com/oss/python/langchain/agents).
 
 Agents use language models to choose a sequence of actions to take.
 
@@ -54,31 +52,33 @@ class AgentAction(Serializable):
     """The input to pass in to the Tool."""
     log: str
     """Additional information to log about the action.
-    This log can be used in a few ways. First, it can be used to audit
-    what exactly the LLM predicted to lead to this (tool, tool_input).
-    Second, it can be used in future iterations to show the LLMs prior
-    thoughts. This is useful when (tool, tool_input) does not contain
-    full information about the LLM prediction (for example, any `thought`
-    before the tool/tool_input)."""
+
+    This log can be used in a few ways. First, it can be used to audit what exactly the
+    LLM predicted to lead to this `(tool, tool_input)`.
+
+    Second, it can be used in future iterations to show the LLMs prior thoughts. This is
+    useful when `(tool, tool_input)` does not contain full information about the LLM
+    prediction (for example, any `thought` before the tool/tool_input).
+    """
     type: Literal["AgentAction"] = "AgentAction"
 
     # Override init to support instantiation by position for backward compat.
     def __init__(self, tool: str, tool_input: str | dict, log: str, **kwargs: Any):
-        """Create an AgentAction.
+        """Create an `AgentAction`.
 
         Args:
             tool: The name of the tool to execute.
-            tool_input: The input to pass in to the Tool.
+            tool_input: The input to pass in to the `Tool`.
             log: Additional information to log about the action.
         """
         super().__init__(tool=tool, tool_input=tool_input, log=log, **kwargs)
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """AgentAction is serializable.
+        """`AgentAction` is serializable.
 
         Returns:
-            True
+            `True`
         """
         return True
 
@@ -100,19 +100,23 @@ class AgentAction(Serializable):
 class AgentActionMessageLog(AgentAction):
     """Representation of an action to be executed by an agent.
 
-    This is similar to AgentAction, but includes a message log consisting of
-    chat messages. This is useful when working with ChatModels, and is used
-    to reconstruct conversation history from the agent's perspective.
+    This is similar to `AgentAction`, but includes a message log consisting of
+    chat messages.
+
+    This is useful when working with `ChatModels`, and is used to reconstruct
+    conversation history from the agent's perspective.
     """
 
     message_log: Sequence[BaseMessage]
-    """Similar to log, this can be used to pass along extra
-    information about what exact messages were predicted by the LLM
-    before parsing out the (tool, tool_input). This is again useful
-    if (tool, tool_input) cannot be used to fully recreate the LLM
-    prediction, and you need that LLM prediction (for future agent iteration).
+    """Similar to log, this can be used to pass along extra information about what exact
+    messages were predicted by the LLM before parsing out the `(tool, tool_input)`.
+
+    This is again useful if `(tool, tool_input)` cannot be used to fully recreate the
+    LLM prediction, and you need that LLM prediction (for future agent iteration).
+
     Compared to `log`, this is useful when the underlying LLM is a
-    chat model (and therefore returns messages rather than a string)."""
+    chat model (and therefore returns messages rather than a string).
+    """
     # Ignoring type because we're overriding the type from AgentAction.
     # And this is the correct thing to do in this case.
     # The type literal is used for serialization purposes.
@@ -120,12 +124,12 @@ class AgentActionMessageLog(AgentAction):
 
 
 class AgentStep(Serializable):
-    """Result of running an AgentAction."""
+    """Result of running an `AgentAction`."""
 
     action: AgentAction
-    """The AgentAction that was executed."""
+    """The `AgentAction` that was executed."""
     observation: Any
-    """The result of the AgentAction."""
+    """The result of the `AgentAction`."""
 
     @property
     def messages(self) -> Sequence[BaseMessage]:
@@ -134,19 +138,22 @@ class AgentStep(Serializable):
 
 
 class AgentFinish(Serializable):
-    """Final return value of an ActionAgent.
+    """Final return value of an `ActionAgent`.
 
-    Agents return an AgentFinish when they have reached a stopping condition.
+    Agents return an `AgentFinish` when they have reached a stopping condition.
     """
 
     return_values: dict
     """Dictionary of return values."""
     log: str
     """Additional information to log about the return value.
+
     This is used to pass along the full LLM prediction, not just the parsed out
-    return value. For example, if the full LLM prediction was
-    `Final Answer: 2` you may want to just return `2` as a return value, but pass
-    along the full string as a `log` (for debugging or observability purposes).
+    return value.
+
+    For example, if the full LLM prediction was `Final Answer: 2` you may want to just
+    return `2` as a return value, but pass along the full string as a `log` (for
+    debugging or observability purposes).
     """
     type: Literal["AgentFinish"] = "AgentFinish"
 
@@ -156,7 +163,7 @@ class AgentFinish(Serializable):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -204,7 +211,7 @@ def _convert_agent_observation_to_messages(
         observation: Observation to convert to a message.
 
     Returns:
-        AIMessage that corresponds to the original tool invocation.
+        `AIMessage` that corresponds to the original tool invocation.
     """
     if isinstance(agent_action, AgentActionMessageLog):
         return [_create_function_message(agent_action, observation)]
@@ -227,7 +234,7 @@ def _create_function_message(
         observation: the result of the tool invocation.
 
     Returns:
-        FunctionMessage that corresponds to the original tool invocation.
+        `FunctionMessage` that corresponds to the original tool invocation.
     """
     if not isinstance(observation, str):
         try:

libs/core/langchain_core/caches.py

@@ -2,8 +2,8 @@
 
 Distinct from provider-based [prompt caching](https://docs.langchain.com/oss/python/langchain/models#prompt-caching).
 
-!!! warning
-    This is a beta feature! Please be wary of deploying experimental code to production
+!!! warning "Beta feature"
+    This is a beta feature. Please be wary of deploying experimental code to production
     unless you've taken appropriate precautions.
 
 A cache is useful for two reasons:
@@ -49,17 +49,18 @@ class BaseCache(ABC):
         """Look up based on `prompt` and `llm_string`.
 
         A cache implementation is expected to generate a key from the 2-tuple
-        of prompt and llm_string (e.g., by concatenating them with a delimiter).
+        of `prompt` and `llm_string` (e.g., by concatenating them with a delimiter).
 
         Args:
             prompt: A string representation of the prompt.
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
-                These invocation parameters are serialized into a string
-                representation.
+
+                These invocation parameters are serialized into a string representation.
 
         Returns:
             On a cache miss, return `None`. On a cache hit, return the cached value.
@@ -78,8 +79,10 @@ class BaseCache(ABC):
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
+
                 These invocation parameters are serialized into a string
                 representation.
             return_val: The value to be cached. The value is a list of `Generation`
@@ -94,15 +97,17 @@ class BaseCache(ABC):
         """Async look up based on `prompt` and `llm_string`.
 
         A cache implementation is expected to generate a key from the 2-tuple
-        of prompt and llm_string (e.g., by concatenating them with a delimiter).
+        of `prompt` and `llm_string` (e.g., by concatenating them with a delimiter).
 
         Args:
             prompt: A string representation of the prompt.
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
+
                 These invocation parameters are serialized into a string
                 representation.
 
@@ -125,8 +130,10 @@ class BaseCache(ABC):
                 In the case of a chat model, the prompt is a non-trivial
                 serialization of the prompt into the language model.
             llm_string: A string representation of the LLM configuration.
+
                 This is used to capture the invocation parameters of the LLM
                 (e.g., model name, temperature, stop tokens, max tokens, etc.).
+
                 These invocation parameters are serialized into a string
                 representation.
             return_val: The value to be cached. The value is a list of `Generation`

libs/core/langchain_core/callbacks/base.py

@@ -5,13 +5,12 @@ from __future__ import annotations
 import logging
 from typing import TYPE_CHECKING, Any
 
-from typing_extensions import Self
-
 if TYPE_CHECKING:
     from collections.abc import Sequence
     from uuid import UUID
 
     from tenacity import RetryCallState
+    from typing_extensions import Self
 
     from langchain_core.agents import AgentAction, AgentFinish
     from langchain_core.documents import Document

libs/core/langchain_core/callbacks/manager.py

@@ -39,7 +39,6 @@ from langchain_core.tracers.context import (
     tracing_v2_callback_var,
 )
 from langchain_core.tracers.langchain import LangChainTracer
-from langchain_core.tracers.schemas import Run
 from langchain_core.tracers.stdout import ConsoleCallbackHandler
 from langchain_core.utils.env import env_var_is_set
 
@@ -52,6 +51,7 @@ if TYPE_CHECKING:
     from langchain_core.documents import Document
     from langchain_core.outputs import ChatGenerationChunk, GenerationChunk, LLMResult
     from langchain_core.runnables.config import RunnableConfig
+    from langchain_core.tracers.schemas import Run
 
 logger = logging.getLogger(__name__)
 
@@ -229,7 +229,24 @@ def shielded(func: Func) -> Func:
 
     @functools.wraps(func)
     async def wrapped(*args: Any, **kwargs: Any) -> Any:
-        return await asyncio.shield(func(*args, **kwargs))
+        # Capture the current context to preserve context variables
+        ctx = copy_context()
+
+        # Create the coroutine
+        coro = func(*args, **kwargs)
+
+        # For Python 3.11+, create task with explicit context
+        # For older versions, fallback to original behavior
+        try:
+            # Create a task with the captured context to preserve context variables
+            task = asyncio.create_task(coro, context=ctx)  # type: ignore[call-arg, unused-ignore]
+            # `call-arg` used to not fail 3.9 or 3.10 tests
+            return await asyncio.shield(task)
+        except TypeError:
+            # Python < 3.11 fallback - create task normally then shield
+            # This won't preserve context perfectly but is better than nothing
+            task = asyncio.create_task(coro)
+            return await asyncio.shield(task)
 
     return cast("Func", wrapped)
 

libs/core/langchain_core/callbacks/usage.py

@@ -24,7 +24,7 @@ class UsageMetadataCallbackHandler(BaseCallbackHandler):
         from langchain_core.callbacks import UsageMetadataCallbackHandler
 
         llm_1 = init_chat_model(model="openai:gpt-4o-mini")
-        llm_2 = init_chat_model(model="anthropic:claude-3-5-haiku-latest")
+        llm_2 = init_chat_model(model="anthropic:claude-3-5-haiku-20241022")
 
         callback = UsageMetadataCallbackHandler()
         result_1 = llm_1.invoke("Hello", config={"callbacks": [callback]})
@@ -43,7 +43,7 @@ class UsageMetadataCallbackHandler(BaseCallbackHandler):
           'input_token_details': {'cache_read': 0, 'cache_creation': 0}}}
         ```
 
-    !!! version-added "Added in version 0.3.49"
+    !!! version-added "Added in `langchain-core` 0.3.49"
 
     """
 
@@ -109,7 +109,7 @@ def get_usage_metadata_callback(
         from langchain_core.callbacks import get_usage_metadata_callback
 
         llm_1 = init_chat_model(model="openai:gpt-4o-mini")
-        llm_2 = init_chat_model(model="anthropic:claude-3-5-haiku-latest")
+        llm_2 = init_chat_model(model="anthropic:claude-3-5-haiku-20241022")
 
         with get_usage_metadata_callback() as cb:
             llm_1.invoke("Hello")
@@ -134,7 +134,7 @@ def get_usage_metadata_callback(
         }
         ```
 
-    !!! version-added "Added in version 0.3.49"
+    !!! version-added "Added in `langchain-core` 0.3.49"
 
     """
     usage_metadata_callback_var: ContextVar[UsageMetadataCallbackHandler | None] = (

libs/core/langchain_core/chat_history.py

@@ -121,7 +121,7 @@ class BaseChatMessageHistory(ABC):
         This method may be deprecated in a future release.
 
         Args:
-            message: The human message to add to the store.
+            message: The `HumanMessage` to add to the store.
         """
         if isinstance(message, HumanMessage):
             self.add_message(message)
@@ -129,7 +129,7 @@ class BaseChatMessageHistory(ABC):
             self.add_message(HumanMessage(content=message))
 
     def add_ai_message(self, message: AIMessage | str) -> None:
-        """Convenience method for adding an AI message string to the store.
+        """Convenience method for adding an `AIMessage` string to the store.
 
         !!! note
             This is a convenience method. Code should favor the bulk `add_messages`
@@ -138,7 +138,7 @@ class BaseChatMessageHistory(ABC):
         This method may be deprecated in a future release.
 
         Args:
-            message: The AI message to add.
+            message: The `AIMessage` to add.
         """
         if isinstance(message, AIMessage):
             self.add_message(message)
@@ -173,7 +173,7 @@ class BaseChatMessageHistory(ABC):
         in an efficient manner to avoid unnecessary round-trips to the underlying store.
 
         Args:
-            messages: A sequence of BaseMessage objects to store.
+            messages: A sequence of `BaseMessage` objects to store.
         """
         for message in messages:
             self.add_message(message)
@@ -182,7 +182,7 @@ class BaseChatMessageHistory(ABC):
         """Async add a list of messages.
 
         Args:
-            messages: A sequence of BaseMessage objects to store.
+            messages: A sequence of `BaseMessage` objects to store.
         """
         await run_in_executor(None, self.add_messages, messages)
 

libs/core/langchain_core/document_loaders/base.py

@@ -27,7 +27,7 @@ class BaseLoader(ABC):  # noqa: B024
     """Interface for Document Loader.
 
     Implementations should implement the lazy-loading method using generators
-    to avoid loading all Documents into memory at once.
+    to avoid loading all documents into memory at once.
 
     `load` is provided just for user convenience and should not be overridden.
     """
@@ -53,9 +53,11 @@ class BaseLoader(ABC):  # noqa: B024
     def load_and_split(
         self, text_splitter: TextSplitter | None = None
     ) -> list[Document]:
-        """Load Documents and split into chunks. Chunks are returned as `Document`.
+        """Load `Document` and split into chunks. Chunks are returned as `Document`.
 
-        Do not override this method. It should be considered to be deprecated!
+        !!! danger
+
+            Do not override this method. It should be considered to be deprecated!
 
         Args:
             text_splitter: `TextSplitter` instance to use for splitting documents.
@@ -135,7 +137,7 @@ class BaseBlobParser(ABC):
         """
 
     def parse(self, blob: Blob) -> list[Document]:
-        """Eagerly parse the blob into a `Document` or `Document` objects.
+        """Eagerly parse the blob into a `Document` or list of `Document` objects.
 
         This is a convenience method for interactive development environment.
 

libs/core/langchain_core/document_loaders/blob_loaders.py

@@ -28,7 +28,7 @@ class BlobLoader(ABC):
     def yield_blobs(
         self,
     ) -> Iterable[Blob]:
-        """A lazy loader for raw data represented by LangChain's Blob object.
+        """A lazy loader for raw data represented by LangChain's `Blob` object.
 
         Returns:
             A generator over blobs

libs/core/langchain_core/document_loaders/langsmith.py

@@ -14,13 +14,13 @@ from langchain_core.documents import Document
 
 
 class LangSmithLoader(BaseLoader):
-    """Load LangSmith Dataset examples as Documents.
+    """Load LangSmith Dataset examples as `Document` objects.
 
-    Loads the example inputs as the Document page content and places the entire example
-    into the Document metadata. This allows you to easily create few-shot example
-    retrievers from the loaded documents.
+    Loads the example inputs as the `Document` page content and places the entire
+    example into the `Document` metadata. This allows you to easily create few-shot
+    example retrievers from the loaded documents.
 
-    ??? note "Lazy load"
+    ??? note "Lazy loading example"
 
         ```python
         from langchain_core.document_loaders import LangSmithLoader
@@ -66,12 +66,11 @@ class LangSmithLoader(BaseLoader):
             format_content: Function for converting the content extracted from the example
                 inputs into a string. Defaults to JSON-encoding the contents.
             example_ids: The IDs of the examples to filter by.
-            as_of: The dataset version tag OR
-                timestamp to retrieve the examples as of.
-                Response examples will only be those that were present at the time
-                of the tagged (or timestamped) version.
+            as_of: The dataset version tag or timestamp to retrieve the examples as of.
+                Response examples will only be those that were present at the time of
+                the tagged (or timestamped) version.
             splits: A list of dataset splits, which are
-                divisions of your dataset such as 'train', 'test', or 'validation'.
+                divisions of your dataset such as `train`, `test`, or `validation`.
                 Returns examples only from the specified splits.
             inline_s3_urls: Whether to inline S3 URLs.
             offset: The offset to start from.

libs/core/langchain_core/documents/__init__.py

@@ -1,7 +1,28 @@
-"""Documents module.
+"""Documents module for data retrieval and processing workflows.
 
-**Document** module is a collection of classes that handle documents
-and their transformations.
+This module provides core abstractions for handling data in retrieval-augmented
+generation (RAG) pipelines, vector stores, and document processing workflows.
+
+!!! warning "Documents vs. message content"
+    This module is distinct from `langchain_core.messages.content`, which provides
+    multimodal content blocks for **LLM chat I/O** (text, images, audio, etc. within
+    messages).
+
+    **Key distinction:**
+
+    - **Documents** (this module): For **data retrieval and processing workflows**
+        - Vector stores, retrievers, RAG pipelines
+        - Text chunking, embedding, and semantic search
+        - Example: Chunks of a PDF stored in a vector database
+
+    - **Content Blocks** (`messages.content`): For **LLM conversational I/O**
+        - Multimodal message content sent to/from models
+        - Tool calls, reasoning, citations within chat
+        - Example: An image sent to a vision model in a chat message (via
+            [`ImageContentBlock`][langchain.messages.ImageContentBlock])
+
+    While both can represent similar data types (text, files), they serve different
+    architectural purposes in LangChain applications.
 """
 
 from typing import TYPE_CHECKING

libs/core/langchain_core/documents/base.py

@@ -1,4 +1,16 @@
-"""Base classes for media and documents."""
+"""Base classes for media and documents.
+
+This module contains core abstractions for **data retrieval and processing workflows**:
+
+- `BaseMedia`: Base class providing `id` and `metadata` fields
+- `Blob`: Raw data loading (files, binary data) - used by document loaders
+- `Document`: Text content for retrieval (RAG, vector stores, semantic search)
+
+!!! note "Not for LLM chat messages"
+    These classes are for data processing pipelines, not LLM I/O. For multimodal
+    content in chat messages (images, audio in conversations), see
+    `langchain.messages` content blocks instead.
+"""
 
 from __future__ import annotations
 
@@ -19,20 +31,18 @@ PathLike = str | PurePath
 
 
 class BaseMedia(Serializable):
-    """Use to represent media content.
+    """Base class for content used in retrieval and data processing workflows.
 
-    Media objects can be used to represent raw data, such as text or binary data.
+    Provides common fields for content that needs to be stored, indexed, or searched.
 
-    LangChain Media objects allow associating metadata and an optional identifier
-    with the content.
-
-    The presence of an ID and metadata make it easier to store, index, and search
-    over the content in a structured way.
+    !!! note
+        For multimodal content in **chat messages** (images, audio sent to/from LLMs),
+        use `langchain.messages` content blocks instead.
     """
 
     # The ID field is optional at the moment.
     # It will likely become required in a future major release after
-    # it has been adopted by enough vectorstore implementations.
+    # it has been adopted by enough VectorStore implementations.
     id: str | None = Field(default=None, coerce_numbers_to_str=True)
     """An optional identifier for the document.
 
@@ -45,71 +55,70 @@ class BaseMedia(Serializable):
 
 
 class Blob(BaseMedia):
-    """Blob represents raw data by either reference or value.
+    """Raw data abstraction for document loading and file processing.
 
-    Provides an interface to materialize the blob in different representations, and
-    help to decouple the development of data loaders from the downstream parsing of
-    the raw data.
+    Represents raw bytes or text, either in-memory or by file reference. Used
+    primarily by document loaders to decouple data loading from parsing.
 
-    Inspired by: https://developer.mozilla.org/en-US/docs/Web/API/Blob
+    Inspired by [Mozilla's `Blob`](https://developer.mozilla.org/en-US/docs/Web/API/Blob)
 
-    Example: Initialize a blob from in-memory data
+    ???+ example "Initialize a blob from in-memory data"
 
-    ```python
-    from langchain_core.documents import Blob
+        ```python
+        from langchain_core.documents import Blob
 
-    blob = Blob.from_data("Hello, world!")
+        blob = Blob.from_data("Hello, world!")
 
-    # Read the blob as a string
-    print(blob.as_string())
+        # Read the blob as a string
+        print(blob.as_string())
 
-    # Read the blob as bytes
-    print(blob.as_bytes())
+        # Read the blob as bytes
+        print(blob.as_bytes())
 
-    # Read the blob as a byte stream
-    with blob.as_bytes_io() as f:
-        print(f.read())
-    ```
+        # Read the blob as a byte stream
+        with blob.as_bytes_io() as f:
+            print(f.read())
+        ```
 
-    Example: Load from memory and specify mime-type and metadata
+    ??? example "Load from memory and specify MIME type and metadata"
 
-    ```python
-    from langchain_core.documents import Blob
+        ```python
+        from langchain_core.documents import Blob
 
-    blob = Blob.from_data(
-        data="Hello, world!",
-        mime_type="text/plain",
-        metadata={"source": "https://example.com"},
-    )
-    ```
+        blob = Blob.from_data(
+            data="Hello, world!",
+            mime_type="text/plain",
+            metadata={"source": "https://example.com"},
+        )
+        ```
 
-    Example: Load the blob from a file
+    ??? example "Load the blob from a file"
 
-    ```python
-    from langchain_core.documents import Blob
+        ```python
+        from langchain_core.documents import Blob
 
-    blob = Blob.from_path("path/to/file.txt")
+        blob = Blob.from_path("path/to/file.txt")
 
-    # Read the blob as a string
-    print(blob.as_string())
+        # Read the blob as a string
+        print(blob.as_string())
 
-    # Read the blob as bytes
-    print(blob.as_bytes())
+        # Read the blob as bytes
+        print(blob.as_bytes())
 
-    # Read the blob as a byte stream
-    with blob.as_bytes_io() as f:
-        print(f.read())
-    ```
+        # Read the blob as a byte stream
+        with blob.as_bytes_io() as f:
+            print(f.read())
+        ```
     """
 
     data: bytes | str | None = None
-    """Raw data associated with the blob."""
+    """Raw data associated with the `Blob`."""
     mimetype: str | None = None
-    """MimeType not to be confused with a file extension."""
+    """MIME type, not to be confused with a file extension."""
     encoding: str = "utf-8"
     """Encoding to use if decoding the bytes into a string.
 
-    Use `utf-8` as default encoding, if decoding to string.
+    Uses `utf-8` as default encoding if decoding to string.
     """
     path: PathLike | None = None
     """Location where the original content was found."""
@@ -123,9 +132,9 @@ class Blob(BaseMedia):
     def source(self) -> str | None:
         """The source location of the blob as string if known otherwise none.
 
-        If a path is associated with the blob, it will default to the path location.
+        If a path is associated with the `Blob`, it will default to the path location.
 
-        Unless explicitly set via a metadata field called `"source"`, in which
+        Unless explicitly set via a metadata field called `'source'`, in which
         case that value will be used instead.
         """
         if self.metadata and "source" in self.metadata:
@@ -211,13 +220,13 @@ class Blob(BaseMedia):
         Args:
             path: Path-like object to file to be read
             encoding: Encoding to use if decoding the bytes into a string
-            mime_type: If provided, will be set as the mime-type of the data
-            guess_type: If `True`, the mimetype will be guessed from the file extension,
-                if a mime-type was not provided
-            metadata: Metadata to associate with the blob
+            mime_type: If provided, will be set as the MIME type of the data
+            guess_type: If `True`, the MIME type will be guessed from the file
+                extension, if a MIME type was not provided
+            metadata: Metadata to associate with the `Blob`
 
         Returns:
-            Blob instance
+            `Blob` instance
         """
         if mime_type is None and guess_type:
             mimetype = mimetypes.guess_type(path)[0] if guess_type else None
@@ -243,17 +252,17 @@ class Blob(BaseMedia):
         path: str | None = None,
         metadata: dict | None = None,
     ) -> Blob:
-        """Initialize the blob from in-memory data.
+        """Initialize the `Blob` from in-memory data.
 
         Args:
-            data: The in-memory data associated with the blob
+            data: The in-memory data associated with the `Blob`
             encoding: Encoding to use if decoding the bytes into a string
-            mime_type: If provided, will be set as the mime-type of the data
+            mime_type: If provided, will be set as the MIME type of the data
             path: If provided, will be set as the source from which the data came
-            metadata: Metadata to associate with the blob
+            metadata: Metadata to associate with the `Blob`
 
         Returns:
-            Blob instance
+            `Blob` instance
         """
         return cls(
             data=data,
@@ -274,6 +283,10 @@ class Blob(BaseMedia):
 class Document(BaseMedia):
     """Class for storing a piece of text and associated metadata.
 
+    !!! note
+        `Document` is for **retrieval workflows**, not chat I/O. For sending text
+        to an LLM in a conversation, use message types from `langchain.messages`.
+
     Example:
         ```python
         from langchain_core.documents import Document
@@ -296,7 +309,7 @@ class Document(BaseMedia):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -309,10 +322,10 @@ class Document(BaseMedia):
         return ["langchain", "schema", "document"]
 
     def __str__(self) -> str:
-        """Override __str__ to restrict it to page_content and metadata.
+        """Override `__str__` to restrict it to page_content and metadata.
 
         Returns:
-            A string representation of the Document.
+            A string representation of the `Document`.
         """
         # The format matches pydantic format for __str__.
         #

libs/core/langchain_core/documents/compressor.py

@@ -21,14 +21,14 @@ class BaseDocumentCompressor(BaseModel, ABC):
 
     This abstraction is primarily used for post-processing of retrieved documents.
 
-    Documents matching a given query are first retrieved.
+    `Document` objects matching a given query are first retrieved.
 
     Then the list of documents can be further processed.
 
     For example, one could re-rank the retrieved documents using an LLM.
 
     !!! note
-        Users should favor using a RunnableLambda instead of sub-classing from this
+        Users should favor using a `RunnableLambda` instead of sub-classing from this
         interface.
 
     """
@@ -43,9 +43,9 @@ class BaseDocumentCompressor(BaseModel, ABC):
         """Compress retrieved documents given the query context.
 
         Args:
-            documents: The retrieved documents.
+            documents: The retrieved `Document` objects.
             query: The query context.
-            callbacks: Optional callbacks to run during compression.
+            callbacks: Optional `Callbacks` to run during compression.
 
         Returns:
             The compressed documents.
@@ -61,9 +61,9 @@ class BaseDocumentCompressor(BaseModel, ABC):
         """Async compress retrieved documents given the query context.
 
         Args:
-            documents: The retrieved documents.
+            documents: The retrieved `Document` objects.
             query: The query context.
-            callbacks: Optional callbacks to run during compression.
+            callbacks: Optional `Callbacks` to run during compression.
 
         Returns:
             The compressed documents.

libs/core/langchain_core/documents/transformers.py

@@ -16,8 +16,8 @@ if TYPE_CHECKING:
 class BaseDocumentTransformer(ABC):
     """Abstract base class for document transformation.
 
-    A document transformation takes a sequence of Documents and returns a
-    sequence of transformed Documents.
+    A document transformation takes a sequence of `Document` objects and returns a
+    sequence of transformed `Document` objects.
 
     Example:
         ```python

libs/core/langchain_core/embeddings/fake.py

@@ -18,7 +18,7 @@ class FakeEmbeddings(Embeddings, BaseModel):
 
     This embedding model creates embeddings by sampling from a normal distribution.
 
-    !!! warning
+    !!! danger "Toy model"
         Do not use this outside of testing, as it is not a real embedding model.
 
     Instantiate:
@@ -73,7 +73,7 @@ class DeterministicFakeEmbedding(Embeddings, BaseModel):
     This embedding model creates embeddings by sampling from a normal distribution
     with a seed based on the hash of the text.
 
-    !!! warning
+    !!! danger "Toy model"
         Do not use this outside of testing, as it is not a real embedding model.
 
     Instantiate:

libs/core/langchain_core/example_selectors/length_based.py

@@ -29,7 +29,7 @@ class LengthBasedExampleSelector(BaseExampleSelector, BaseModel):
     max_length: int = 2048
     """Max length for the prompt, beyond which examples are cut."""
 
-    example_text_lengths: list[int] = Field(default_factory=list)  # :meta private:
+    example_text_lengths: list[int] = Field(default_factory=list)
     """Length of each example."""
 
     def add_example(self, example: dict[str, str]) -> None:

libs/core/langchain_core/example_selectors/semantic_similarity.py

@@ -41,7 +41,7 @@ class _VectorStoreExampleSelector(BaseExampleSelector, BaseModel, ABC):
     """Optional keys to filter input to. If provided, the search is based on
     the input variables instead of all variables."""
     vectorstore_kwargs: dict[str, Any] | None = None
-    """Extra arguments passed to similarity_search function of the vectorstore."""
+    """Extra arguments passed to similarity_search function of the `VectorStore`."""
 
     model_config = ConfigDict(
         arbitrary_types_allowed=True,
@@ -159,7 +159,7 @@ class SemanticSimilarityExampleSelector(_VectorStoreExampleSelector):
                 instead of all variables.
             example_keys: If provided, keys to filter examples to.
             vectorstore_kwargs: Extra arguments passed to similarity_search function
-                of the vectorstore.
+                of the `VectorStore`.
             vectorstore_cls_kwargs: optional kwargs containing url for vector store
 
         Returns:
@@ -203,7 +203,7 @@ class SemanticSimilarityExampleSelector(_VectorStoreExampleSelector):
                 instead of all variables.
             example_keys: If provided, keys to filter examples to.
             vectorstore_kwargs: Extra arguments passed to similarity_search function
-                of the vectorstore.
+                of the `VectorStore`.
             vectorstore_cls_kwargs: optional kwargs containing url for vector store
 
         Returns:
@@ -286,12 +286,12 @@ class MaxMarginalRelevanceExampleSelector(_VectorStoreExampleSelector):
             embeddings: An initialized embedding API interface, e.g. OpenAIEmbeddings().
             vectorstore_cls: A vector store DB interface class, e.g. FAISS.
             k: Number of examples to select.
-            fetch_k: Number of Documents to fetch to pass to MMR algorithm.
+            fetch_k: Number of `Document` objects to fetch to pass to MMR algorithm.
             input_keys: If provided, the search is based on the input variables
                 instead of all variables.
             example_keys: If provided, keys to filter examples to.
             vectorstore_kwargs: Extra arguments passed to similarity_search function
-                of the vectorstore.
+                of the `VectorStore`.
             vectorstore_cls_kwargs: optional kwargs containing url for vector store
 
         Returns:
@@ -333,12 +333,12 @@ class MaxMarginalRelevanceExampleSelector(_VectorStoreExampleSelector):
             embeddings: An initialized embedding API interface, e.g. OpenAIEmbeddings().
             vectorstore_cls: A vector store DB interface class, e.g. FAISS.
             k: Number of examples to select.
-            fetch_k: Number of Documents to fetch to pass to MMR algorithm.
+            fetch_k: Number of `Document` objects to fetch to pass to MMR algorithm.
             input_keys: If provided, the search is based on the input variables
                 instead of all variables.
             example_keys: If provided, keys to filter examples to.
             vectorstore_kwargs: Extra arguments passed to similarity_search function
-                of the vectorstore.
+                of the `VectorStore`.
             vectorstore_cls_kwargs: optional kwargs containing url for vector store
 
         Returns:

libs/core/langchain_core/exceptions.py

@@ -16,9 +16,10 @@ class OutputParserException(ValueError, LangChainException):  # noqa: N818
     """Exception that output parsers should raise to signify a parsing error.
 
     This exists to differentiate parsing errors from other code or execution errors
-    that also may arise inside the output parser. `OutputParserException` will be
-    available to catch and handle in ways to fix the parsing error, while other
-    errors will be raised.
+    that also may arise inside the output parser.
+
+    `OutputParserException` will be available to catch and handle in ways to fix the
+    parsing error, while other errors will be raised.
     """
 
     def __init__(
@@ -32,18 +33,19 @@ class OutputParserException(ValueError, LangChainException):  # noqa: N818
 
         Args:
             error: The error that's being re-raised or an error message.
-            observation: String explanation of error which can be passed to a
-                model to try and remediate the issue.
+            observation: String explanation of error which can be passed to a model to
+                try and remediate the issue.
             llm_output: String model output which is error-ing.
 
             send_to_llm: Whether to send the observation and llm_output back to an Agent
                 after an `OutputParserException` has been raised.
+
                 This gives the underlying model driving the agent the context that the
                 previous output was improperly structured, in the hopes that it will
                 update the output to the correct format.
 
         Raises:
-            ValueError: If `send_to_llm` is True but either observation or
+            ValueError: If `send_to_llm` is `True` but either observation or
                 `llm_output` are not provided.
         """
         if isinstance(error, str):
@@ -66,11 +68,11 @@ class ErrorCode(Enum):
     """Error codes."""
 
     INVALID_PROMPT_INPUT = "INVALID_PROMPT_INPUT"
-    INVALID_TOOL_RESULTS = "INVALID_TOOL_RESULTS"
+    INVALID_TOOL_RESULTS = "INVALID_TOOL_RESULTS"  # Used in JS; not Py (yet)
     MESSAGE_COERCION_FAILURE = "MESSAGE_COERCION_FAILURE"
-    MODEL_AUTHENTICATION = "MODEL_AUTHENTICATION"
-    MODEL_NOT_FOUND = "MODEL_NOT_FOUND"
-    MODEL_RATE_LIMIT = "MODEL_RATE_LIMIT"
+    MODEL_AUTHENTICATION = "MODEL_AUTHENTICATION"  # Used in JS; not Py (yet)
+    MODEL_NOT_FOUND = "MODEL_NOT_FOUND"  # Used in JS; not Py (yet)
+    MODEL_RATE_LIMIT = "MODEL_RATE_LIMIT"  # Used in JS; not Py (yet)
     OUTPUT_PARSING_FAILURE = "OUTPUT_PARSING_FAILURE"
 
 
@@ -86,6 +88,6 @@ def create_message(*, message: str, error_code: ErrorCode) -> str:
     """
     return (
         f"{message}\n"
-        "For troubleshooting, visit: https://python.langchain.com/docs/"
-        f"troubleshooting/errors/{error_code.value} "
+        "For troubleshooting, visit: https://docs.langchain.com/oss/python/langchain"
+        f"/errors/{error_code.value} "
     )

libs/core/langchain_core/indexing/__init__.py

@@ -1,7 +1,7 @@
 """Code to help indexing data into a vectorstore.
 
 This package contains helper logic to help deal with indexing data into
-a vectorstore while avoiding duplicated content and over-writing content
+a `VectorStore` while avoiding duplicated content and over-writing content
 if it's unchanged.
 """
 

libs/core/langchain_core/indexing/api.py

@@ -6,16 +6,9 @@ import hashlib
 import json
 import uuid
 import warnings
-from collections.abc import (
-    AsyncIterable,
-    AsyncIterator,
-    Callable,
-    Iterable,
-    Iterator,
-    Sequence,
-)
 from itertools import islice
 from typing import (
+    TYPE_CHECKING,
     Any,
     Literal,
     TypedDict,
@@ -29,6 +22,16 @@ from langchain_core.exceptions import LangChainException
 from langchain_core.indexing.base import DocumentIndex, RecordManager
 from langchain_core.vectorstores import VectorStore
 
+if TYPE_CHECKING:
+    from collections.abc import (
+        AsyncIterable,
+        AsyncIterator,
+        Callable,
+        Iterable,
+        Iterator,
+        Sequence,
+    )
+
 # Magic UUID to use as a namespace for hashing.
 # Used to try and generate a unique UUID for each document
 # from hashing the document content and metadata.
@@ -298,48 +301,49 @@ def index(
     For the time being, documents are indexed using their hashes, and users
     are not able to specify the uid of the document.
 
-    !!! warning "Behavior changed in 0.3.25"
+    !!! warning "Behavior changed in `langchain-core` 0.3.25"
+
         Added `scoped_full` cleanup mode.
 
     !!! warning
 
         * In full mode, the loader should be returning
-          the entire dataset, and not just a subset of the dataset.
-          Otherwise, the auto_cleanup will remove documents that it is not
-          supposed to.
+            the entire dataset, and not just a subset of the dataset.
+            Otherwise, the auto_cleanup will remove documents that it is not
+            supposed to.
         * In incremental mode, if documents associated with a particular
-          source id appear across different batches, the indexing API
-          will do some redundant work. This will still result in the
-          correct end state of the index, but will unfortunately not be
-          100% efficient. For example, if a given document is split into 15
-          chunks, and we index them using a batch size of 5, we'll have 3 batches
-          all with the same source id. In general, to avoid doing too much
-          redundant work select as big a batch size as possible.
+            source id appear across different batches, the indexing API
+            will do some redundant work. This will still result in the
+            correct end state of the index, but will unfortunately not be
+            100% efficient. For example, if a given document is split into 15
+            chunks, and we index them using a batch size of 5, we'll have 3 batches
+            all with the same source id. In general, to avoid doing too much
+            redundant work select as big a batch size as possible.
         * The `scoped_full` mode is suitable if determining an appropriate batch size
-          is challenging or if your data loader cannot return the entire dataset at
-          once. This mode keeps track of source IDs in memory, which should be fine
-          for most use cases. If your dataset is large (10M+ docs), you will likely
-          need to parallelize the indexing process regardless.
+            is challenging or if your data loader cannot return the entire dataset at
+            once. This mode keeps track of source IDs in memory, which should be fine
+            for most use cases. If your dataset is large (10M+ docs), you will likely
+            need to parallelize the indexing process regardless.
 
     Args:
         docs_source: Data loader or iterable of documents to index.
         record_manager: Timestamped set to keep track of which documents were
             updated.
-        vector_store: VectorStore or DocumentIndex to index the documents into.
+        vector_store: `VectorStore` or DocumentIndex to index the documents into.
         batch_size: Batch size to use when indexing.
         cleanup: How to handle clean up of documents.
 
             - incremental: Cleans up all documents that haven't been updated AND
-              that are associated with source ids that were seen during indexing.
-              Clean up is done continuously during indexing helping to minimize the
-              probability of users seeing duplicated content.
+                that are associated with source IDs that were seen during indexing.
+                Clean up is done continuously during indexing helping to minimize the
+                probability of users seeing duplicated content.
             - full: Delete all documents that have not been returned by the loader
-              during this run of indexing.
-              Clean up runs after all documents have been indexed.
-              This means that users may see duplicated content during indexing.
+                during this run of indexing.
+                Clean up runs after all documents have been indexed.
+                This means that users may see duplicated content during indexing.
             - scoped_full: Similar to Full, but only deletes all documents
-              that haven't been updated AND that are associated with
-              source ids that were seen during indexing.
+                that haven't been updated AND that are associated with
+                source IDs that were seen during indexing.
             - None: Do not delete any documents.
         source_id_key: Optional key that helps identify the original source
             of the document.
@@ -349,7 +353,7 @@ def index(
         key_encoder: Hashing algorithm to use for hashing the document content and
             metadata. Options include "blake2b", "sha256", and "sha512".
 
-            !!! version-added "Added in version 0.3.66"
+            !!! version-added "Added in `langchain-core` 0.3.66"
 
         key_encoder: Hashing algorithm to use for hashing the document.
             If not provided, a default encoder using SHA-1 will be used.
@@ -363,10 +367,10 @@ def index(
             When changing the key encoder, you must change the
             index as well to avoid duplicated documents in the cache.
         upsert_kwargs: Additional keyword arguments to pass to the add_documents
-            method of the VectorStore or the upsert method of the DocumentIndex.
+            method of the `VectorStore` or the upsert method of the DocumentIndex.
             For example, you can use this to specify a custom vector_field:
             upsert_kwargs={"vector_field": "embedding"}
-            !!! version-added "Added in version 0.3.10"
+            !!! version-added "Added in `langchain-core` 0.3.10"
 
     Returns:
         Indexing result which contains information about how many documents
@@ -375,10 +379,10 @@ def index(
     Raises:
         ValueError: If cleanup mode is not one of 'incremental', 'full' or None
         ValueError: If cleanup mode is incremental and source_id_key is None.
-        ValueError: If vectorstore does not have
+        ValueError: If `VectorStore` does not have
             "delete" and "add_documents" required methods.
         ValueError: If source_id_key is not None, but is not a string or callable.
-        TypeError: If `vectorstore` is not a VectorStore or a DocumentIndex.
+        TypeError: If `vectorstore` is not a `VectorStore` or a DocumentIndex.
         AssertionError: If `source_id` is None when cleanup mode is incremental.
             (should be unreachable code).
     """
@@ -415,7 +419,7 @@ def index(
                 raise ValueError(msg)
 
         if type(destination).delete == VectorStore.delete:
-            # Checking if the vectorstore has overridden the default delete method
+            # Checking if the VectorStore has overridden the default delete method
             # implementation which just raises a NotImplementedError
             msg = "Vectorstore has not implemented the delete method"
             raise ValueError(msg)
@@ -466,11 +470,11 @@ def index(
         ]
 
         if cleanup in {"incremental", "scoped_full"}:
-            # source ids are required.
+            # Source IDs are required.
             for source_id, hashed_doc in zip(source_ids, hashed_docs, strict=False):
                 if source_id is None:
                     msg = (
-                        f"Source ids are required when cleanup mode is "
+                        f"Source IDs are required when cleanup mode is "
                         f"incremental or scoped_full. "
                         f"Document that starts with "
                         f"content: {hashed_doc.page_content[:100]} "
@@ -479,7 +483,7 @@ def index(
                     raise ValueError(msg)
                 if cleanup == "scoped_full":
                     scoped_full_cleanup_source_ids.add(source_id)
-            # source ids cannot be None after for loop above.
+            # Source IDs cannot be None after for loop above.
             source_ids = cast("Sequence[str]", source_ids)
 
         exists_batch = record_manager.exists(
@@ -538,7 +542,7 @@ def index(
         # If source IDs are provided, we can do the deletion incrementally!
         if cleanup == "incremental":
             # Get the uids of the documents that were not returned by the loader.
-            # mypy isn't good enough to determine that source ids cannot be None
+            # mypy isn't good enough to determine that source IDs cannot be None
             # here due to a check that's happening above, so we check again.
             for source_id in source_ids:
                 if source_id is None:
@@ -636,48 +640,49 @@ async def aindex(
     For the time being, documents are indexed using their hashes, and users
     are not able to specify the uid of the document.
 
-    !!! warning "Behavior changed in 0.3.25"
+    !!! warning "Behavior changed in `langchain-core` 0.3.25"
+
         Added `scoped_full` cleanup mode.
 
     !!! warning
 
         * In full mode, the loader should be returning
-          the entire dataset, and not just a subset of the dataset.
-          Otherwise, the auto_cleanup will remove documents that it is not
-          supposed to.
+            the entire dataset, and not just a subset of the dataset.
+            Otherwise, the auto_cleanup will remove documents that it is not
+            supposed to.
         * In incremental mode, if documents associated with a particular
-          source id appear across different batches, the indexing API
-          will do some redundant work. This will still result in the
-          correct end state of the index, but will unfortunately not be
-          100% efficient. For example, if a given document is split into 15
-          chunks, and we index them using a batch size of 5, we'll have 3 batches
-          all with the same source id. In general, to avoid doing too much
-          redundant work select as big a batch size as possible.
+            source id appear across different batches, the indexing API
+            will do some redundant work. This will still result in the
+            correct end state of the index, but will unfortunately not be
+            100% efficient. For example, if a given document is split into 15
+            chunks, and we index them using a batch size of 5, we'll have 3 batches
+            all with the same source id. In general, to avoid doing too much
+            redundant work select as big a batch size as possible.
         * The `scoped_full` mode is suitable if determining an appropriate batch size
-          is challenging or if your data loader cannot return the entire dataset at
-          once. This mode keeps track of source IDs in memory, which should be fine
-          for most use cases. If your dataset is large (10M+ docs), you will likely
-          need to parallelize the indexing process regardless.
+            is challenging or if your data loader cannot return the entire dataset at
+            once. This mode keeps track of source IDs in memory, which should be fine
+            for most use cases. If your dataset is large (10M+ docs), you will likely
+            need to parallelize the indexing process regardless.
 
     Args:
         docs_source: Data loader or iterable of documents to index.
         record_manager: Timestamped set to keep track of which documents were
             updated.
-        vector_store: VectorStore or DocumentIndex to index the documents into.
+        vector_store: `VectorStore` or DocumentIndex to index the documents into.
         batch_size: Batch size to use when indexing.
         cleanup: How to handle clean up of documents.
 
             - incremental: Cleans up all documents that haven't been updated AND
-              that are associated with source ids that were seen during indexing.
-              Clean up is done continuously during indexing helping to minimize the
-              probability of users seeing duplicated content.
+                that are associated with source IDs that were seen during indexing.
+                Clean up is done continuously during indexing helping to minimize the
+                probability of users seeing duplicated content.
             - full: Delete all documents that have not been returned by the loader
-              during this run of indexing.
-              Clean up runs after all documents have been indexed.
-              This means that users may see duplicated content during indexing.
+                during this run of indexing.
+                Clean up runs after all documents have been indexed.
+                This means that users may see duplicated content during indexing.
             - scoped_full: Similar to Full, but only deletes all documents
-              that haven't been updated AND that are associated with
-              source ids that were seen during indexing.
+                that haven't been updated AND that are associated with
+                source IDs that were seen during indexing.
             - None: Do not delete any documents.
         source_id_key: Optional key that helps identify the original source
             of the document.
@@ -687,7 +692,7 @@ async def aindex(
         key_encoder: Hashing algorithm to use for hashing the document content and
             metadata. Options include "blake2b", "sha256", and "sha512".
 
-            !!! version-added "Added in version 0.3.66"
+            !!! version-added "Added in `langchain-core` 0.3.66"
 
         key_encoder: Hashing algorithm to use for hashing the document.
             If not provided, a default encoder using SHA-1 will be used.
@@ -701,10 +706,10 @@ async def aindex(
             When changing the key encoder, you must change the
             index as well to avoid duplicated documents in the cache.
         upsert_kwargs: Additional keyword arguments to pass to the add_documents
-            method of the VectorStore or the upsert method of the DocumentIndex.
+            method of the `VectorStore` or the upsert method of the DocumentIndex.
             For example, you can use this to specify a custom vector_field:
             upsert_kwargs={"vector_field": "embedding"}
-            !!! version-added "Added in version 0.3.10"
+            !!! version-added "Added in `langchain-core` 0.3.10"
 
     Returns:
         Indexing result which contains information about how many documents
@@ -713,10 +718,10 @@ async def aindex(
     Raises:
         ValueError: If cleanup mode is not one of 'incremental', 'full' or None
         ValueError: If cleanup mode is incremental and source_id_key is None.
-        ValueError: If vectorstore does not have
+        ValueError: If `VectorStore` does not have
             "adelete" and "aadd_documents" required methods.
         ValueError: If source_id_key is not None, but is not a string or callable.
-        TypeError: If `vector_store` is not a VectorStore or DocumentIndex.
+        TypeError: If `vector_store` is not a `VectorStore` or DocumentIndex.
         AssertionError: If `source_id_key` is None when cleanup mode is
             incremental or `scoped_full` (should be unreachable).
     """
@@ -757,7 +762,7 @@ async def aindex(
             type(destination).adelete == VectorStore.adelete
             and type(destination).delete == VectorStore.delete
         ):
-            # Checking if the vectorstore has overridden the default adelete or delete
+            # Checking if the VectorStore has overridden the default adelete or delete
             # methods implementation which just raises a NotImplementedError
             msg = "Vectorstore has not implemented the adelete or delete method"
             raise ValueError(msg)
@@ -815,11 +820,11 @@ async def aindex(
         ]
 
         if cleanup in {"incremental", "scoped_full"}:
-            # If the cleanup mode is incremental, source ids are required.
+            # If the cleanup mode is incremental, source IDs are required.
             for source_id, hashed_doc in zip(source_ids, hashed_docs, strict=False):
                 if source_id is None:
                     msg = (
-                        f"Source ids are required when cleanup mode is "
+                        f"Source IDs are required when cleanup mode is "
                         f"incremental or scoped_full. "
                         f"Document that starts with "
                         f"content: {hashed_doc.page_content[:100]} "
@@ -828,7 +833,7 @@ async def aindex(
                     raise ValueError(msg)
                 if cleanup == "scoped_full":
                     scoped_full_cleanup_source_ids.add(source_id)
-            # source ids cannot be None after for loop above.
+            # Source IDs cannot be None after for loop above.
             source_ids = cast("Sequence[str]", source_ids)
 
         exists_batch = await record_manager.aexists(
@@ -888,7 +893,7 @@ async def aindex(
         if cleanup == "incremental":
             # Get the uids of the documents that were not returned by the loader.
 
-            # mypy isn't good enough to determine that source ids cannot be None
+            # mypy isn't good enough to determine that source IDs cannot be None
             # here due to a check that's happening above, so we check again.
             for source_id in source_ids:
                 if source_id is None:

libs/core/langchain_core/indexing/base.py

@@ -25,7 +25,7 @@ class RecordManager(ABC):
     The record manager abstraction is used by the langchain indexing API.
 
     The record manager keeps track of which documents have been
-    written into a vectorstore and when they were written.
+    written into a `VectorStore` and when they were written.
 
     The indexing API computes hashes for each document and stores the hash
     together with the write time and the source id in the record manager.
@@ -37,7 +37,7 @@ class RecordManager(ABC):
     already been indexed, and to only index new documents.
 
     The main benefit of this abstraction is that it works across many vectorstores.
-    To be supported, a vectorstore needs to only support the ability to add and
+    To be supported, a `VectorStore` needs to only support the ability to add and
     delete documents by ID. Using the record manager, the indexing API will
     be able to delete outdated documents and avoid redundant indexing of documents
     that have already been indexed.
@@ -45,13 +45,13 @@ class RecordManager(ABC):
     The main constraints of this abstraction are:
 
     1. It relies on the time-stamps to determine which documents have been
-       indexed and which have not. This means that the time-stamps must be
-       monotonically increasing. The timestamp should be the timestamp
-       as measured by the server to minimize issues.
+        indexed and which have not. This means that the time-stamps must be
+        monotonically increasing. The timestamp should be the timestamp
+        as measured by the server to minimize issues.
     2. The record manager is currently implemented separately from the
-       vectorstore, which means that the overall system becomes distributed
-       and may create issues with consistency. For example, writing to
-       record manager succeeds, but corresponding writing to vectorstore fails.
+        vectorstore, which means that the overall system becomes distributed
+        and may create issues with consistency. For example, writing to
+        record manager succeeds, but corresponding writing to `VectorStore` fails.
     """
 
     def __init__(
@@ -460,7 +460,7 @@ class UpsertResponse(TypedDict):
 class DeleteResponse(TypedDict, total=False):
     """A generic response for delete operation.
 
-    The fields in this response are optional and whether the vectorstore
+    The fields in this response are optional and whether the `VectorStore`
     returns them or not is up to the implementation.
     """
 
@@ -518,7 +518,7 @@ class DocumentIndex(BaseRetriever):
         if it is provided. If the ID is not provided, the upsert method is free
         to generate an ID for the content.
 
-        When an ID is specified and the content already exists in the vectorstore,
+        When an ID is specified and the content already exists in the `VectorStore`,
         the upsert method should update the content with the new data. If the content
         does not exist, the upsert method should add the item to the `VectorStore`.
 
@@ -528,20 +528,20 @@ class DocumentIndex(BaseRetriever):
 
         Returns:
             A response object that contains the list of IDs that were
-            successfully added or updated in the vectorstore and the list of IDs that
+            successfully added or updated in the `VectorStore` and the list of IDs that
             failed to be added or updated.
         """
 
     async def aupsert(
         self, items: Sequence[Document], /, **kwargs: Any
     ) -> UpsertResponse:
-        """Add or update documents in the vectorstore. Async version of upsert.
+        """Add or update documents in the `VectorStore`. Async version of `upsert`.
 
         The upsert functionality should utilize the ID field of the item
         if it is provided. If the ID is not provided, the upsert method is free
         to generate an ID for the item.
 
-        When an ID is specified and the item already exists in the vectorstore,
+        When an ID is specified and the item already exists in the `VectorStore`,
         the upsert method should update the item with the new data. If the item
         does not exist, the upsert method should add the item to the `VectorStore`.
 
@@ -551,7 +551,7 @@ class DocumentIndex(BaseRetriever):
 
         Returns:
             A response object that contains the list of IDs that were
-            successfully added or updated in the vectorstore and the list of IDs that
+            successfully added or updated in the `VectorStore` and the list of IDs that
             failed to be added or updated.
         """
         return await run_in_executor(
@@ -568,7 +568,7 @@ class DocumentIndex(BaseRetriever):
         Calling delete without any input parameters should raise a ValueError!
 
         Args:
-            ids: List of ids to delete.
+            ids: List of IDs to delete.
             **kwargs: Additional keyword arguments. This is up to the implementation.
                 For example, can include an option to delete the entire index,
                 or else issue a non-blocking delete etc.
@@ -586,7 +586,7 @@ class DocumentIndex(BaseRetriever):
         Calling adelete without any input parameters should raise a ValueError!
 
         Args:
-            ids: List of ids to delete.
+            ids: List of IDs to delete.
             **kwargs: Additional keyword arguments. This is up to the implementation.
                 For example, can include an option to delete the entire index.
 

libs/core/langchain_core/indexing/in_memory.py

@@ -62,10 +62,10 @@ class InMemoryDocumentIndex(DocumentIndex):
         """Delete by IDs.
 
         Args:
-            ids: List of ids to delete.
+            ids: List of IDs to delete.
 
         Raises:
-            ValueError: If ids is None.
+            ValueError: If IDs is None.
 
         Returns:
             A response object that contains the list of IDs that were successfully

libs/core/langchain_core/language_models/__init__.py

@@ -6,12 +6,13 @@ LangChain has two main classes to work with language models: chat models and
 **Chat models**
 
 Language models that use a sequence of messages as inputs and return chat messages
-as outputs (as opposed to using plain text). Chat models support the assignment of
-distinct roles to conversation messages, helping to distinguish messages from the AI,
-users, and instructions such as system messages.
+as outputs (as opposed to using plain text).
 
-The key abstraction for chat models is `BaseChatModel`. Implementations
-should inherit from this class.
+Chat models support the assignment of distinct roles to conversation messages, helping
+to distinguish messages from the AI, users, and instructions such as system messages.
+
+The key abstraction for chat models is `BaseChatModel`. Implementations should inherit
+from this class.
 
 See existing [chat model integrations](https://docs.langchain.com/oss/python/integrations/chat).
 
@@ -52,6 +53,10 @@ if TYPE_CHECKING:
         ParrotFakeChatModel,
     )
     from langchain_core.language_models.llms import LLM, BaseLLM
+    from langchain_core.language_models.model_profile import (
+        ModelProfile,
+        ModelProfileRegistry,
+    )
 
 __all__ = (
     "LLM",
@@ -67,6 +72,8 @@ __all__ = (
     "LanguageModelInput",
     "LanguageModelLike",
     "LanguageModelOutput",
+    "ModelProfile",
+    "ModelProfileRegistry",
     "ParrotFakeChatModel",
     "SimpleChatModel",
     "get_tokenizer",
@@ -89,6 +96,8 @@ _dynamic_imports = {
     "GenericFakeChatModel": "fake_chat_models",
     "ParrotFakeChatModel": "fake_chat_models",
     "LLM": "llms",
+    "ModelProfile": "model_profile",
+    "ModelProfileRegistry": "model_profile",
     "BaseLLM": "llms",
     "is_openai_data_block": "_utils",
 }

libs/core/langchain_core/language_models/_utils.py

@@ -139,7 +139,8 @@ def _normalize_messages(
             directly; this may change in the future
     - LangChain v0 standard content blocks for backward compatibility
 
-    !!! warning "Behavior changed in 1.0.0"
+    !!! warning "Behavior changed in `langchain-core` 1.0.0"
+
         In previous versions, this function returned messages in LangChain v0 format.
         Now, it returns messages in LangChain v1 format, which upgraded chat models now
         expect to receive when passing back in message history. For backward

libs/core/langchain_core/language_models/base.py

@@ -131,14 +131,19 @@ class BaseLanguageModel(
 
     Caching is not currently supported for streaming methods of models.
     """
+
     verbose: bool = Field(default_factory=_get_verbosity, exclude=True, repr=False)
     """Whether to print out response text."""
+
     callbacks: Callbacks = Field(default=None, exclude=True)
     """Callbacks to add to the run trace."""
+
     tags: list[str] | None = Field(default=None, exclude=True)
     """Tags to add to the run trace."""
+
     metadata: dict[str, Any] | None = Field(default=None, exclude=True)
     """Metadata to add to the run trace."""
+
     custom_get_token_ids: Callable[[str], list[int]] | None = Field(
         default=None, exclude=True
     )
@@ -195,15 +200,22 @@ class BaseLanguageModel(
             type (e.g., pure text completion models vs chat models).
 
         Args:
-            prompts: List of `PromptValue` objects. A `PromptValue` is an object that
-                can be converted to match the format of any language model (string for
-                pure text generation models and `BaseMessage` objects for chat models).
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            prompts: List of `PromptValue` objects.
+
+                A `PromptValue` is an object that can be converted to match the format
+                of any language model (string for pure text generation models and
+                `BaseMessage` objects for chat models).
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generation` objects for
@@ -232,15 +244,22 @@ class BaseLanguageModel(
             type (e.g., pure text completion models vs chat models).
 
         Args:
-            prompts: List of `PromptValue` objects. A `PromptValue` is an object that
-                can be converted to match the format of any language model (string for
-                pure text generation models and `BaseMessage` objects for chat models).
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            prompts: List of `PromptValue` objects.
+
+                A `PromptValue` is an object that can be converted to match the format
+                of any language model (string for pure text generation models and
+                `BaseMessage` objects for chat models).
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generation` objects for
@@ -262,13 +281,13 @@ class BaseLanguageModel(
         return self.lc_attributes
 
     def get_token_ids(self, text: str) -> list[int]:
-        """Return the ordered ids of the tokens in a text.
+        """Return the ordered IDs of the tokens in a text.
 
         Args:
             text: The string input to tokenize.
 
         Returns:
-            A list of ids corresponding to the tokens in the text, in order they occur
+            A list of IDs corresponding to the tokens in the text, in order they occur
                 in the text.
         """
         if self.custom_get_token_ids is not None:
@@ -280,6 +299,9 @@ class BaseLanguageModel(
 
         Useful for checking if an input fits in a model's context window.
 
+        This should be overridden by model-specific implementations to provide accurate
+        token counts via model-specific tokenizers.
+
         Args:
             text: The string input to tokenize.
 
@@ -298,9 +320,17 @@ class BaseLanguageModel(
 
         Useful for checking if an input fits in a model's context window.
 
+        This should be overridden by model-specific implementations to provide accurate
+        token counts via model-specific tokenizers.
+
         !!! note
-            The base implementation of `get_num_tokens_from_messages` ignores tool
-            schemas.
+
+            * The base implementation of `get_num_tokens_from_messages` ignores tool
+                schemas.
+            * The base implementation of `get_num_tokens_from_messages` adds additional
+                prefixes to messages in represent user roles, which will add to the
+                overall token count. Model-specific implementations may choose to
+                handle this differently.
 
         Args:
             messages: The message inputs to tokenize.

libs/core/langchain_core/language_models/chat_models.py

@@ -33,6 +33,7 @@ from langchain_core.language_models.base import (
     LangSmithParams,
     LanguageModelInput,
 )
+from langchain_core.language_models.model_profile import ModelProfile
 from langchain_core.load import dumpd, dumps
 from langchain_core.messages import (
     AIMessage,
@@ -88,7 +89,10 @@ def _generate_response_from_error(error: BaseException) -> list[ChatGeneration]:
             try:
                 metadata["body"] = response.json()
             except Exception:
-                metadata["body"] = getattr(response, "text", None)
+                try:
+                    metadata["body"] = getattr(response, "text", None)
+                except Exception:
+                    metadata["body"] = None
         if hasattr(response, "headers"):
             try:
                 metadata["headers"] = dict(response.headers)
@@ -329,10 +333,25 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
     [`langchain-openai`](https://pypi.org/project/langchain-openai)) can also use this
     field to roll out new content formats in a backward-compatible way.
 
-    !!! version-added "Added in version 1.0"
+    !!! version-added "Added in `langchain-core` 1.0.0"
 
     """
 
+    profile: ModelProfile | None = Field(default=None, exclude=True)
+    """Profile detailing model capabilities.
+
+    !!! warning "Beta feature"
+        This is a beta feature. The format of model profiles is subject to change.
+
+    If not specified, automatically loaded from the provider package on initialization
+    if data is available.
+
+    Example profile data includes context window sizes, supported modalities, or support
+    for tool calling, structured output, and other features.
+
+    !!! version-added "Added in `langchain-core` 1.1.0"
+    """
+
     model_config = ConfigDict(
         arbitrary_types_allowed=True,
     )
@@ -842,16 +861,21 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
 
         Args:
             messages: List of list of messages.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: The tags to apply.
             metadata: The metadata to apply.
             run_name: The name of the run.
             run_id: The ID of the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generations` for each
@@ -960,16 +984,21 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
 
         Args:
             messages: List of list of messages.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: The tags to apply.
             metadata: The metadata to apply.
             run_name: The name of the run.
             run_id: The ID of the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generations` for each
@@ -1502,10 +1531,10 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
         Args:
             schema: The output schema. Can be passed in as:
 
-                - an OpenAI function/tool schema,
-                - a JSON Schema,
-                - a `TypedDict` class,
-                - or a Pydantic class.
+                - An OpenAI function/tool schema,
+                - A JSON Schema,
+                - A `TypedDict` class,
+                - Or a Pydantic class.
 
                 If `schema` is a Pydantic class then the model output will be a
                 Pydantic instance of that class, and the model-generated fields will be
@@ -1517,11 +1546,15 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
                 when specifying a Pydantic or `TypedDict` class.
 
             include_raw:
-                If `False` then only the parsed structured output is returned. If
-                an error occurs during model output parsing it will be raised. If `True`
-                then both the raw model response (a `BaseMessage`) and the parsed model
-                response will be returned. If an error occurs during output parsing it
-                will be caught and returned as well.
+                If `False` then only the parsed structured output is returned.
+
+                If an error occurs during model output parsing it will be raised.
+
+                If `True` then both the raw model response (a `BaseMessage`) and the
+                parsed model response will be returned.
+
+                If an error occurs during output parsing it will be caught and returned
+                as well.
 
                 The final output is always a `dict` with keys `'raw'`, `'parsed'`, and
                 `'parsing_error'`.
@@ -1599,7 +1632,7 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
         # }
         ```
 
-        Example: `dict` schema (`include_raw=False`):
+        Example: Dictionary schema (`include_raw=False`):
 
         ```python
         from pydantic import BaseModel
@@ -1626,8 +1659,9 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
         # }
         ```
 
-        !!! warning "Behavior changed in 0.2.26"
-            Added support for TypedDict class.
+        !!! warning "Behavior changed in `langchain-core` 0.2.26"
+
+            Added support for `TypedDict` class.
 
         """  # noqa: E501
         _ = kwargs.pop("method", None)
@@ -1726,9 +1760,12 @@ def _gen_info_and_msg_metadata(
     }
 
 
+_MAX_CLEANUP_DEPTH = 100
+
+
 def _cleanup_llm_representation(serialized: Any, depth: int) -> None:
     """Remove non-serializable objects from a serialized object."""
-    if depth > 100:  # Don't cooperate for pathological cases
+    if depth > _MAX_CLEANUP_DEPTH:  # Don't cooperate for pathological cases
         return
 
     if not isinstance(serialized, dict):

libs/core/langchain_core/language_models/fake_chat_models.py

@@ -1,4 +1,4 @@
-"""Fake chat model for testing purposes."""
+"""Fake chat models for testing purposes."""
 
 import asyncio
 import re

libs/core/langchain_core/language_models/llms.py

@@ -1,4 +1,7 @@
-"""Base interface for large language models to expose."""
+"""Base interface for traditional large language models (LLMs) to expose.
+
+These are traditionally older models (newer models generally are chat models).
+"""
 
 from __future__ import annotations
 
@@ -648,9 +651,12 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: The prompts to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
 
         Returns:
@@ -668,9 +674,12 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: The prompts to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
 
         Returns:
@@ -702,11 +711,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Yields:
             Generation chunks.
@@ -728,11 +740,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Yields:
             Generation chunks.
@@ -843,10 +858,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: List of string prompts.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: List of tags to associate with each prompt. If provided, the length
                 of the list must match the length of the prompts list.
             metadata: List of metadata dictionaries to associate with each prompt. If
@@ -856,8 +875,9 @@ class BaseLLM(BaseLanguageModel[str], ABC):
                 length of the list must match the length of the prompts list.
             run_id: List of run IDs to associate with each prompt. If provided, the
                 length of the list must match the length of the prompts list.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Raises:
             ValueError: If prompts is not a list.
@@ -1113,10 +1133,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: List of string prompts.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: List of tags to associate with each prompt. If provided, the length
                 of the list must match the length of the prompts list.
             metadata: List of metadata dictionaries to associate with each prompt. If
@@ -1126,8 +1150,9 @@ class BaseLLM(BaseLanguageModel[str], ABC):
                 length of the list must match the length of the prompts list.
             run_id: List of run IDs to associate with each prompt. If provided, the
                 length of the list must match the length of the prompts list.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Raises:
             ValueError: If the length of `callbacks`, `tags`, `metadata`, or
@@ -1391,11 +1416,6 @@ class LLM(BaseLLM):
         `astream` will use `_astream` if provided, otherwise it will implement
         a fallback behavior that will use `_stream` if `_stream` is implemented,
         and use `_acall` if `_stream` is not implemented.
-
-    Please see the following guide for more information on how to
-    implement a custom LLM:
-
-    https://python.langchain.com/docs/how_to/custom_llm/
     """
 
     @abstractmethod
@@ -1412,12 +1432,16 @@ class LLM(BaseLLM):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             The model output as a string. SHOULD NOT include the prompt.
@@ -1438,12 +1462,16 @@ class LLM(BaseLLM):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             The model output as a string. SHOULD NOT include the prompt.

libs/core/langchain_core/language_models/model_profile.py

@@ -0,0 +1,84 @@
+"""Model profile types and utilities."""
+
+from typing_extensions import TypedDict
+
+
+class ModelProfile(TypedDict, total=False):
+    """Model profile.
+
+    !!! warning "Beta feature"
+        This is a beta feature. The format of model profiles is subject to change.
+
+    Provides information about chat model capabilities, such as context window sizes
+    and supported features.
+    """
+
+    # --- Input constraints ---
+
+    max_input_tokens: int
+    """Maximum context window (tokens)"""
+
+    image_inputs: bool
+    """Whether image inputs are supported."""
+    # TODO: add more detail about formats?
+
+    image_url_inputs: bool
+    """Whether [image URL inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    pdf_inputs: bool
+    """Whether [PDF inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+    # TODO: add more detail about formats? e.g. bytes or base64
+
+    audio_inputs: bool
+    """Whether [audio inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+    # TODO: add more detail about formats? e.g. bytes or base64
+
+    video_inputs: bool
+    """Whether [video inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+    # TODO: add more detail about formats? e.g. bytes or base64
+
+    image_tool_message: bool
+    """Whether images can be included in tool messages."""
+
+    pdf_tool_message: bool
+    """Whether PDFs can be included in tool messages."""
+
+    # --- Output constraints ---
+
+    max_output_tokens: int
+    """Maximum output tokens"""
+
+    reasoning_output: bool
+    """Whether the model supports [reasoning / chain-of-thought](https://docs.langchain.com/oss/python/langchain/models#reasoning)"""
+
+    image_outputs: bool
+    """Whether [image outputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    audio_outputs: bool
+    """Whether [audio outputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    video_outputs: bool
+    """Whether [video outputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    # --- Tool calling ---
+    tool_calling: bool
+    """Whether the model supports [tool calling](https://docs.langchain.com/oss/python/langchain/models#tool-calling)"""
+
+    tool_choice: bool
+    """Whether the model supports [tool choice](https://docs.langchain.com/oss/python/langchain/models#forcing-tool-calls)"""
+
+    # --- Structured output ---
+    structured_output: bool
+    """Whether the model supports a native [structured output](https://docs.langchain.com/oss/python/langchain/models#structured-outputs)
+    feature"""
+
+
+ModelProfileRegistry = dict[str, ModelProfile]
+"""Registry mapping model identifiers or names to their ModelProfile."""

libs/core/langchain_core/load/dump.py

@@ -17,7 +17,7 @@ def default(obj: Any) -> Any:
         obj: The object to serialize to json if it is a Serializable object.
 
     Returns:
-        A json serializable object or a SerializedNotImplemented object.
+        A JSON serializable object or a SerializedNotImplemented object.
     """
     if isinstance(obj, Serializable):
         return obj.to_json()

libs/core/langchain_core/load/load.py

@@ -61,13 +61,15 @@ class Reviver:
         """Initialize the reviver.
 
         Args:
-            secrets_map: A map of secrets to load. If a secret is not found in
-                the map, it will be loaded from the environment if `secrets_from_env`
-                is True.
+            secrets_map: A map of secrets to load.
+
+                If a secret is not found in the map, it will be loaded from the
+                environment if `secrets_from_env` is `True`.
             valid_namespaces: A list of additional namespaces (modules)
                 to allow to be deserialized.
             secrets_from_env: Whether to load secrets from the environment.
             additional_import_mappings: A dictionary of additional namespace mappings
+
                 You can use this to override default mappings or add new mappings.
             ignore_unserializable_fields: Whether to ignore unserializable fields.
         """
@@ -195,13 +197,15 @@ def loads(
 
     Args:
         text: The string to load.
-        secrets_map: A map of secrets to load. If a secret is not found in
-            the map, it will be loaded from the environment if `secrets_from_env`
-            is True.
+        secrets_map: A map of secrets to load.
+
+            If a secret is not found in the map, it will be loaded from the environment
+            if `secrets_from_env` is `True`.
         valid_namespaces: A list of additional namespaces (modules)
             to allow to be deserialized.
         secrets_from_env: Whether to load secrets from the environment.
         additional_import_mappings: A dictionary of additional namespace mappings
+
             You can use this to override default mappings or add new mappings.
         ignore_unserializable_fields: Whether to ignore unserializable fields.
 
@@ -237,13 +241,15 @@ def load(
 
     Args:
         obj: The object to load.
-        secrets_map: A map of secrets to load. If a secret is not found in
-            the map, it will be loaded from the environment if `secrets_from_env`
-            is True.
+        secrets_map: A map of secrets to load.
+
+            If a secret is not found in the map, it will be loaded from the environment
+            if `secrets_from_env` is `True`.
         valid_namespaces: A list of additional namespaces (modules)
             to allow to be deserialized.
         secrets_from_env: Whether to load secrets from the environment.
         additional_import_mappings: A dictionary of additional namespace mappings
+
             You can use this to override default mappings or add new mappings.
         ignore_unserializable_fields: Whether to ignore unserializable fields.
 

libs/core/langchain_core/load/serializable.py

@@ -97,11 +97,14 @@ class Serializable(BaseModel, ABC):
         by default. This is to prevent accidental serialization of objects that should
         not be serialized.
     - `get_lc_namespace`: Get the namespace of the LangChain object.
+
         During deserialization, this namespace is used to identify
         the correct class to instantiate.
+
         Please see the `Reviver` class in `langchain_core.load.load` for more details.
         During deserialization an additional mapping is handle classes that have moved
         or been renamed across package versions.
+
     - `lc_secrets`: A map of constructor argument names to secret ids.
     - `lc_attributes`: List of additional attribute names that should be included
         as part of the serialized representation.
@@ -194,7 +197,7 @@ class Serializable(BaseModel, ABC):
             ValueError: If the class has deprecated attributes.
 
         Returns:
-            A json serializable object or a `SerializedNotImplemented` object.
+            A JSON serializable object or a `SerializedNotImplemented` object.
         """
         if not self.is_lc_serializable():
             return self.to_json_not_implemented()

libs/core/langchain_core/messages/__init__.py

@@ -9,6 +9,9 @@ if TYPE_CHECKING:
     from langchain_core.messages.ai import (
         AIMessage,
         AIMessageChunk,
+        InputTokenDetails,
+        OutputTokenDetails,
+        UsageMetadata,
     )
     from langchain_core.messages.base import (
         BaseMessage,
@@ -87,10 +90,12 @@ __all__ = (
     "HumanMessage",
     "HumanMessageChunk",
     "ImageContentBlock",
+    "InputTokenDetails",
     "InvalidToolCall",
     "MessageLikeRepresentation",
     "NonStandardAnnotation",
     "NonStandardContentBlock",
+    "OutputTokenDetails",
     "PlainTextContentBlock",
     "ReasoningContentBlock",
     "RemoveMessage",
@@ -104,6 +109,7 @@ __all__ = (
     "ToolCallChunk",
     "ToolMessage",
     "ToolMessageChunk",
+    "UsageMetadata",
     "VideoContentBlock",
     "_message_from_dict",
     "convert_to_messages",
@@ -145,6 +151,7 @@ _dynamic_imports = {
     "HumanMessageChunk": "human",
     "NonStandardAnnotation": "content",
     "NonStandardContentBlock": "content",
+    "OutputTokenDetails": "ai",
     "PlainTextContentBlock": "content",
     "ReasoningContentBlock": "content",
     "RemoveMessage": "modifier",
@@ -154,12 +161,14 @@ _dynamic_imports = {
     "SystemMessage": "system",
     "SystemMessageChunk": "system",
     "ImageContentBlock": "content",
+    "InputTokenDetails": "ai",
     "InvalidToolCall": "tool",
     "TextContentBlock": "content",
     "ToolCall": "tool",
     "ToolCallChunk": "tool",
     "ToolMessage": "tool",
     "ToolMessageChunk": "tool",
+    "UsageMetadata": "ai",
     "VideoContentBlock": "content",
     "AnyMessage": "utils",
     "MessageLikeRepresentation": "utils",

libs/core/langchain_core/messages/ai.py

@@ -48,10 +48,10 @@ class InputTokenDetails(TypedDict, total=False):
         }
         ```
 
-    !!! version-added "Added in version 0.3.9"
-
     May also hold extra provider-specific keys.
 
+    !!! version-added "Added in `langchain-core` 0.3.9"
+
     """
 
     audio: int
@@ -83,7 +83,9 @@ class OutputTokenDetails(TypedDict, total=False):
         }
         ```
 
-    !!! version-added "Added in version 0.3.9"
+    May also hold extra provider-specific keys.
+
+    !!! version-added "Added in `langchain-core` 0.3.9"
 
     """
 
@@ -121,9 +123,15 @@ class UsageMetadata(TypedDict):
         }
         ```
 
-    !!! warning "Behavior changed in 0.3.9"
+    !!! warning "Behavior changed in `langchain-core` 0.3.9"
+
         Added `input_token_details` and `output_token_details`.
 
+    !!! note "LangSmith SDK"
+
+        The LangSmith SDK also has a `UsageMetadata` class. While the two share fields,
+        LangSmith's `UsageMetadata` has additional fields to capture cost information
+        used by the LangSmith platform.
     """
 
     input_tokens: int
@@ -131,7 +139,7 @@ class UsageMetadata(TypedDict):
     output_tokens: int
     """Count of output (or completion) tokens. Sum of all output token types."""
     total_tokens: int
-    """Total token count. Sum of input_tokens + output_tokens."""
+    """Total token count. Sum of `input_tokens` + `output_tokens`."""
     input_token_details: NotRequired[InputTokenDetails]
     """Breakdown of input token counts.
 
@@ -141,7 +149,6 @@ class UsageMetadata(TypedDict):
     """Breakdown of output token counts.
 
     Does *not* need to sum to full output token count. Does *not* need to have all keys.
-
     """
 
 
@@ -153,7 +160,6 @@ class AIMessage(BaseMessage):
     This message represents the output of the model and consists of both
     the raw output as returned by the model and standardized fields
     (e.g., tool calls, usage metadata) added by the LangChain framework.
-
     """
 
     tool_calls: list[ToolCall] = []
@@ -651,13 +657,13 @@ def add_ai_message_chunks(
             chunk_id = id_
             break
     else:
-        # second pass: prefer lc_run-* ids over lc_* ids
+        # second pass: prefer lc_run-* IDs over lc_* IDs
         for id_ in candidates:
             if id_ and id_.startswith(LC_ID_PREFIX):
                 chunk_id = id_
                 break
         else:
-            # third pass: take any remaining id (auto-generated lc_* ids)
+            # third pass: take any remaining ID (auto-generated lc_* IDs)
             for id_ in candidates:
                 if id_:
                     chunk_id = id_

libs/core/langchain_core/messages/base.py

@@ -5,11 +5,9 @@ from __future__ import annotations
 from typing import TYPE_CHECKING, Any, cast, overload
 
 from pydantic import ConfigDict, Field
-from typing_extensions import Self
 
 from langchain_core._api.deprecation import warn_deprecated
 from langchain_core.load.serializable import Serializable
-from langchain_core.messages import content as types
 from langchain_core.utils import get_bolded_text
 from langchain_core.utils._merge import merge_dicts, merge_lists
 from langchain_core.utils.interactive_env import is_interactive_env
@@ -17,6 +15,9 @@ from langchain_core.utils.interactive_env import is_interactive_env
 if TYPE_CHECKING:
     from collections.abc import Sequence
 
+    from typing_extensions import Self
+
+    from langchain_core.messages import content as types
     from langchain_core.prompts.chat import ChatPromptTemplate
 
 
@@ -93,6 +94,10 @@ class BaseMessage(Serializable):
     """Base abstract message class.
 
     Messages are the inputs and outputs of a chat model.
+
+    Examples include [`HumanMessage`][langchain.messages.HumanMessage],
+    [`AIMessage`][langchain.messages.AIMessage], and
+    [`SystemMessage`][langchain.messages.SystemMessage].
     """
 
     content: str | list[str | dict]
@@ -195,7 +200,7 @@ class BaseMessage(Serializable):
     def content_blocks(self) -> list[types.ContentBlock]:
         r"""Load content blocks from the message content.
 
-        !!! version-added "Added in version 1.0.0"
+        !!! version-added "Added in `langchain-core` 1.0.0"
 
         """
         # Needed here to avoid circular import, as these classes import BaseMessages

libs/core/langchain_core/messages/block_translators/__init__.py

@@ -12,10 +12,11 @@ the implementation in `BaseMessage`.
 
 from __future__ import annotations
 
-from collections.abc import Callable
 from typing import TYPE_CHECKING
 
 if TYPE_CHECKING:
+    from collections.abc import Callable
+
     from langchain_core.messages import AIMessage, AIMessageChunk
     from langchain_core.messages import content as types
 

libs/core/langchain_core/messages/block_translators/google_genai.py

@@ -368,7 +368,7 @@ def _convert_to_v1_from_genai(message: AIMessage) -> list[types.ContentBlock]:
                     else:
                         # Assume it's raw base64 without data URI
                         try:
-                            # Validate base64 and decode for mime type detection
+                            # Validate base64 and decode for MIME type detection
                             decoded_bytes = base64.b64decode(url, validate=True)
 
                             image_url_b64_block = {
@@ -379,7 +379,7 @@ def _convert_to_v1_from_genai(message: AIMessage) -> list[types.ContentBlock]:
                             try:
                                 import filetype  # type: ignore[import-not-found] # noqa: PLC0415
 
-                                # Guess mime type based on file bytes
+                                # Guess MIME type based on file bytes
                                 mime_type = None
                                 kind = filetype.guess(decoded_bytes)
                                 if kind:
@@ -458,6 +458,8 @@ def _convert_to_v1_from_genai(message: AIMessage) -> list[types.ContentBlock]:
                 if outcome is not None:
                     server_tool_result_block["extras"]["outcome"] = outcome
                 converted_blocks.append(server_tool_result_block)
+            elif item_type == "text":
+                converted_blocks.append(cast("types.TextContentBlock", item))
             else:
                 # Unknown type, preserve as non-standard
                 converted_blocks.append({"type": "non_standard", "value": item})

libs/core/langchain_core/messages/block_translators/openai.py

@@ -4,7 +4,6 @@ from __future__ import annotations
 
 import json
 import warnings
-from collections.abc import Iterable
 from typing import TYPE_CHECKING, Any, Literal, cast
 
 from langchain_core.language_models._utils import (
@@ -14,6 +13,8 @@ from langchain_core.language_models._utils import (
 from langchain_core.messages import content as types
 
 if TYPE_CHECKING:
+    from collections.abc import Iterable
+
     from langchain_core.messages import AIMessage, AIMessageChunk
 
 

libs/core/langchain_core/messages/content.py

@@ -644,7 +644,7 @@ class AudioContentBlock(TypedDict):
 
 
 class PlainTextContentBlock(TypedDict):
-    """Plaintext data (e.g., from a document).
+    """Plaintext data (e.g., from a `.txt` or `.md` document).
 
     !!! note
         A `PlainTextContentBlock` existed in `langchain-core<1.0.0`. Although the
@@ -654,7 +654,7 @@ class PlainTextContentBlock(TypedDict):
 
     !!! note
         Title and context are optional fields that may be passed to the model. See
-        Anthropic [example](https://docs.claude.com/en/docs/build-with-claude/citations#citable-vs-non-citable-content).
+        Anthropic [example](https://platform.claude.com/docs/en/build-with-claude/citations#citable-vs-non-citable-content).
 
     !!! note "Factory function"
         `create_plaintext_block` may also be used as a factory to create a
@@ -767,7 +767,7 @@ class FileContentBlock(TypedDict):
 
 
 class NonStandardContentBlock(TypedDict):
-    """Provider-specific data.
+    """Provider-specific content data.
 
     This block contains data for which there is not yet a standard type.
 
@@ -802,7 +802,7 @@ class NonStandardContentBlock(TypedDict):
     """
 
     value: dict[str, Any]
-    """Provider-specific data."""
+    """Provider-specific content data."""
 
     index: NotRequired[int | str]
     """Index of block in aggregate response. Used during streaming."""
@@ -867,7 +867,7 @@ def _get_data_content_block_types() -> tuple[str, ...]:
     Example: ("image", "video", "audio", "text-plain", "file")
 
     Note that old style multimodal blocks type literals with new style blocks.
-    Speficially, "image", "audio", and "file".
+    Specifically, "image", "audio", and "file".
 
     See the docstring of `_normalize_messages` in `language_models._utils` for details.
     """
@@ -906,7 +906,7 @@ def is_data_content_block(block: dict) -> bool:
 
         # 'text' is checked to support v0 PlainTextContentBlock types
         # We must guard against new style TextContentBlock which also has 'text' `type`
-        # by ensuring the presense of `source_type`
+        # by ensuring the presence of `source_type`
         if block["type"] == "text" and "source_type" not in block:  # noqa: SIM103  # This is more readable
             return False
 
@@ -1399,7 +1399,7 @@ def create_non_standard_block(
     """Create a `NonStandardContentBlock`.
 
     Args:
-        value: Provider-specific data.
+        value: Provider-specific content data.
         id: Content block identifier. Generated automatically if not provided.
         index: Index of block in aggregate response. Used during streaming.
 

libs/core/langchain_core/messages/utils.py

@@ -86,7 +86,7 @@ AnyMessage = Annotated[
     | Annotated[ToolMessageChunk, Tag(tag="ToolMessageChunk")],
     Field(discriminator=Discriminator(_get_type)),
 ]
-""""A type representing any defined `Message` or `MessageChunk` type."""
+"""A type representing any defined `Message` or `MessageChunk` type."""
 
 
 def get_buffer_string(
@@ -328,12 +328,16 @@ def _convert_to_message(message: MessageLikeRepresentation) -> BaseMessage:
     """
     if isinstance(message, BaseMessage):
         message_ = message
-    elif isinstance(message, str):
-        message_ = _create_message_from_message_type("human", message)
-    elif isinstance(message, Sequence) and len(message) == 2:
-        # mypy doesn't realise this can't be a string given the previous branch
-        message_type_str, template = message  # type: ignore[misc]
-        message_ = _create_message_from_message_type(message_type_str, template)
+    elif isinstance(message, Sequence):
+        if isinstance(message, str):
+            message_ = _create_message_from_message_type("human", message)
+        else:
+            try:
+                message_type_str, template = message
+            except ValueError as e:
+                msg = "Message as a sequence must be (role string, template)"
+                raise NotImplementedError(msg) from e
+            message_ = _create_message_from_message_type(message_type_str, template)
     elif isinstance(message, dict):
         msg_kwargs = message.copy()
         try:
@@ -734,8 +738,10 @@ def trim_messages(
             Set to `len` to count the number of **messages** in the chat history.
 
             !!! note
+
                 Use `count_tokens_approximately` to get fast, approximate token
                 counts.
+
                 This is recommended for using `trim_messages` on the hot path, where
                 exact token counting is not necessary.
 
@@ -1025,18 +1031,18 @@ def convert_to_openai_messages(
         messages: Message-like object or iterable of objects whose contents are
             in OpenAI, Anthropic, Bedrock Converse, or VertexAI formats.
         text_format: How to format string or text block contents:
-                - `'string'`:
-                    If a message has a string content, this is left as a string. If
-                    a message has content blocks that are all of type `'text'`, these
-                    are joined with a newline to make a single string. If a message has
-                    content blocks and at least one isn't of type `'text'`, then
-                    all blocks are left as dicts.
-                - `'block'`:
-                    If a message has a string content, this is turned into a list
-                    with a single content block of type `'text'`. If a message has
-                    content blocks these are left as is.
-        include_id: Whether to include message ids in the openai messages, if they
-                    are present in the source messages.
+            - `'string'`:
+                If a message has a string content, this is left as a string. If
+                a message has content blocks that are all of type `'text'`, these
+                are joined with a newline to make a single string. If a message has
+                content blocks and at least one isn't of type `'text'`, then
+                all blocks are left as dicts.
+            - `'block'`:
+                If a message has a string content, this is turned into a list
+                with a single content block of type `'text'`. If a message has
+                content blocks these are left as is.
+        include_id: Whether to include message IDs in the openai messages, if they
+            are present in the source messages.
 
     Raises:
         ValueError: if an unrecognized `text_format` is specified, or if a message
@@ -1097,7 +1103,7 @@ def convert_to_openai_messages(
         # ]
         ```
 
-    !!! version-added "Added in version 0.3.11"
+    !!! version-added "Added in `langchain-core` 0.3.11"
 
     """  # noqa: E501
     if text_format not in {"string", "block"}:
@@ -1697,7 +1703,7 @@ def count_tokens_approximately(
     Warning:
         This function does not currently support counting image tokens.
 
-    !!! version-added "Added in version 0.3.46"
+    !!! version-added "Added in `langchain-core` 0.3.46"
 
     """
     token_count = 0.0

libs/core/langchain_core/output_parsers/__init__.py

@@ -1,4 +1,20 @@
-"""**OutputParser** classes parse the output of an LLM call."""
+"""`OutputParser` classes parse the output of an LLM call into structured data.
+
+!!! tip "Structured output"
+
+    Output parsers emerged as an early solution to the challenge of obtaining structured
+    output from LLMs.
+
+    Today, most LLMs support [structured output](https://docs.langchain.com/oss/python/langchain/models#structured-outputs)
+    natively. In such cases, using output parsers may be unnecessary, and you should
+    leverage the model's built-in capabilities for structured output. Refer to the
+    [documentation of your chosen model](https://docs.langchain.com/oss/python/integrations/providers/overview)
+    for guidance on how to achieve structured output directly.
+
+    Output parsers remain valuable when working with models that do not support
+    structured output natively, or when you require additional processing or validation
+    of the model's output beyond its inherent capabilities.
+"""
 
 from typing import TYPE_CHECKING
 

libs/core/langchain_core/output_parsers/base.py

@@ -135,6 +135,9 @@ class BaseOutputParser(
 
     Example:
         ```python
+        # Implement a simple boolean output parser
+
+
         class BooleanOutputParser(BaseOutputParser[bool]):
             true_val: str = "YES"
             false_val: str = "NO"

libs/core/langchain_core/output_parsers/format_instructions.py

@@ -1,11 +1,16 @@
 """Format instructions."""
 
-JSON_FORMAT_INSTRUCTIONS = """The output should be formatted as a JSON instance that conforms to the JSON schema below.
+JSON_FORMAT_INSTRUCTIONS = """STRICT OUTPUT FORMAT:
+- Return only the JSON value that conforms to the schema. Do not include any additional text, explanations, headings, or separators.
+- Do not wrap the JSON in Markdown or code fences (no ``` or ```json).
+- Do not prepend or append any text (e.g., do not write "Here is the JSON:").
+- The response must be a single top-level JSON value exactly as required by the schema (object/array/etc.), with no trailing commas or comments.
 
-As an example, for the schema {{"properties": {{"foo": {{"title": "Foo", "description": "a list of strings", "type": "array", "items": {{"type": "string"}}}}}}, "required": ["foo"]}}
-the object {{"foo": ["bar", "baz"]}} is a well-formatted instance of the schema. The object {{"properties": {{"foo": ["bar", "baz"]}}}} is not well-formatted.
+The output should be formatted as a JSON instance that conforms to the JSON schema below.
 
-Here is the output schema:
+As an example, for the schema {{"properties": {{"foo": {{"title": "Foo", "description": "a list of strings", "type": "array", "items": {{"type": "string"}}}}}}, "required": ["foo"]}} the object {{"foo": ["bar", "baz"]}} is a well-formatted instance of the schema. The object {{"properties": {{"foo": ["bar", "baz"]}}}} is not well-formatted.
+
+Here is the output schema (shown in a code block for readability only — do not include any backticks or Markdown in your output):
 ```
 {schema}
 ```"""  # noqa: E501

libs/core/langchain_core/output_parsers/json.py

@@ -31,11 +31,14 @@ TBaseModel = TypeVar("TBaseModel", bound=PydanticBaseModel)
 class JsonOutputParser(BaseCumulativeTransformOutputParser[Any]):
     """Parse the output of an LLM call to a JSON object.
 
+    Probably the most reliable output parser for getting structured data that does *not*
+    use function calling.
+
     When used in streaming mode, it will yield partial JSON objects containing
     all the keys that have been returned so far.
 
-    In streaming, if `diff` is set to `True`, yields JSONPatch operations
-    describing the difference between the previous and the current object.
+    In streaming, if `diff` is set to `True`, yields JSONPatch operations describing the
+    difference between the previous and the current object.
     """
 
     pydantic_object: Annotated[type[TBaseModel] | None, SkipValidation()] = None  # type: ignore[valid-type]

libs/core/langchain_core/output_parsers/list.py

@@ -41,7 +41,7 @@ def droplastn(
 
 
 class ListOutputParser(BaseTransformOutputParser[list[str]]):
-    """Parse the output of an LLM call to a list."""
+    """Parse the output of a model to a list."""
 
     @property
     def _type(self) -> str:
@@ -74,30 +74,30 @@ class ListOutputParser(BaseTransformOutputParser[list[str]]):
         buffer = ""
         for chunk in input:
             if isinstance(chunk, BaseMessage):
-                # extract text
+                # Extract text
                 chunk_content = chunk.content
                 if not isinstance(chunk_content, str):
                     continue
                 buffer += chunk_content
             else:
-                # add current chunk to buffer
+                # Add current chunk to buffer
                 buffer += chunk
-            # parse buffer into a list of parts
+            # Parse buffer into a list of parts
             try:
                 done_idx = 0
-                # yield only complete parts
+                # Yield only complete parts
                 for m in droplastn(self.parse_iter(buffer), 1):
                     done_idx = m.end()
                     yield [m.group(1)]
                 buffer = buffer[done_idx:]
             except NotImplementedError:
                 parts = self.parse(buffer)
-                # yield only complete parts
+                # Yield only complete parts
                 if len(parts) > 1:
                     for part in parts[:-1]:
                         yield [part]
                     buffer = parts[-1]
-        # yield the last part
+        # Yield the last part
         for part in self.parse(buffer):
             yield [part]
 
@@ -108,40 +108,40 @@ class ListOutputParser(BaseTransformOutputParser[list[str]]):
         buffer = ""
         async for chunk in input:
             if isinstance(chunk, BaseMessage):
-                # extract text
+                # Extract text
                 chunk_content = chunk.content
                 if not isinstance(chunk_content, str):
                     continue
                 buffer += chunk_content
             else:
-                # add current chunk to buffer
+                # Add current chunk to buffer
                 buffer += chunk
-            # parse buffer into a list of parts
+            # Parse buffer into a list of parts
             try:
                 done_idx = 0
-                # yield only complete parts
+                # Yield only complete parts
                 for m in droplastn(self.parse_iter(buffer), 1):
                     done_idx = m.end()
                     yield [m.group(1)]
                 buffer = buffer[done_idx:]
             except NotImplementedError:
                 parts = self.parse(buffer)
-                # yield only complete parts
+                # Yield only complete parts
                 if len(parts) > 1:
                     for part in parts[:-1]:
                         yield [part]
                     buffer = parts[-1]
-        # yield the last part
+        # Yield the last part
         for part in self.parse(buffer):
             yield [part]
 
 
 class CommaSeparatedListOutputParser(ListOutputParser):
-    """Parse the output of an LLM call to a comma-separated list."""
+    """Parse the output of a model to a comma-separated list."""
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -177,7 +177,7 @@ class CommaSeparatedListOutputParser(ListOutputParser):
             )
             return [item for sublist in reader for item in sublist]
         except csv.Error:
-            # keep old logic for backup
+            # Keep old logic for backup
             return [part.strip() for part in text.split(",")]
 
     @property

libs/core/langchain_core/output_parsers/openai_tools.py

@@ -15,7 +15,11 @@ from langchain_core.messages.tool import tool_call as create_tool_call
 from langchain_core.output_parsers.transform import BaseCumulativeTransformOutputParser
 from langchain_core.outputs import ChatGeneration, Generation
 from langchain_core.utils.json import parse_partial_json
-from langchain_core.utils.pydantic import TypeBaseModel
+from langchain_core.utils.pydantic import (
+    TypeBaseModel,
+    is_pydantic_v1_subclass,
+    is_pydantic_v2_subclass,
+)
 
 logger = logging.getLogger(__name__)
 
@@ -224,7 +228,7 @@ class JsonOutputKeyToolsParser(JsonOutputToolsParser):
             result: The result of the LLM call.
             partial: Whether to parse partial JSON.
                 If `True`, the output will be a JSON object containing
-                all the keys that have been returned so far.
+                    all the keys that have been returned so far.
                 If `False`, the output will be the full JSON object.
 
         Raises:
@@ -307,7 +311,7 @@ class PydanticToolsParser(JsonOutputToolsParser):
             result: The result of the LLM call.
             partial: Whether to parse partial JSON.
                 If `True`, the output will be a JSON object containing
-                all the keys that have been returned so far.
+                    all the keys that have been returned so far.
                 If `False`, the output will be the full JSON object.
 
         Returns:
@@ -323,7 +327,15 @@ class PydanticToolsParser(JsonOutputToolsParser):
             return None if self.first_tool_only else []
 
         json_results = [json_results] if self.first_tool_only else json_results
-        name_dict = {tool.__name__: tool for tool in self.tools}
+        name_dict_v2: dict[str, TypeBaseModel] = {
+            tool.model_config.get("title") or tool.__name__: tool
+            for tool in self.tools
+            if is_pydantic_v2_subclass(tool)
+        }
+        name_dict_v1: dict[str, TypeBaseModel] = {
+            tool.__name__: tool for tool in self.tools if is_pydantic_v1_subclass(tool)
+        }
+        name_dict: dict[str, TypeBaseModel] = {**name_dict_v2, **name_dict_v1}
         pydantic_objects = []
         for res in json_results:
             if not isinstance(res["args"], dict):

libs/core/langchain_core/output_parsers/pydantic.py

@@ -37,7 +37,7 @@ class PydanticOutputParser(JsonOutputParser, Generic[TBaseModel]):
     def _parser_exception(
         self, e: Exception, json_object: dict
     ) -> OutputParserException:
-        json_string = json.dumps(json_object)
+        json_string = json.dumps(json_object, ensure_ascii=False)
         name = self.pydantic_object.__name__
         msg = f"Failed to parse {name} from completion {json_string}. Got: {e}"
         return OutputParserException(msg, llm_output=json_string)
@@ -86,7 +86,7 @@ class PydanticOutputParser(JsonOutputParser, Generic[TBaseModel]):
             The format instructions for the JSON output.
         """
         # Copy schema to avoid altering original Pydantic schema.
-        schema = dict(self.pydantic_object.model_json_schema().items())
+        schema = dict(self._get_schema(self.pydantic_object).items())
 
         # Remove extraneous fields.
         reduced_schema = schema

libs/core/langchain_core/output_parsers/string.py

@@ -6,14 +6,14 @@ from langchain_core.output_parsers.transform import BaseTransformOutputParser
 
 
 class StrOutputParser(BaseTransformOutputParser[str]):
-    """OutputParser that parses LLMResult into the top likely string."""
+    """OutputParser that parses `LLMResult` into the top likely string."""
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """StrOutputParser is serializable.
+        """`StrOutputParser` is serializable.
 
         Returns:
-            True
+            `True`
         """
         return True
 

libs/core/langchain_core/output_parsers/xml.py

@@ -43,19 +43,19 @@ class _StreamingParser:
     """Streaming parser for XML.
 
     This implementation is pulled into a class to avoid implementation
-    drift between transform and atransform of the XMLOutputParser.
+    drift between transform and atransform of the `XMLOutputParser`.
     """
 
     def __init__(self, parser: Literal["defusedxml", "xml"]) -> None:
         """Initialize the streaming parser.
 
         Args:
-            parser: Parser to use for XML parsing. Can be either 'defusedxml' or 'xml'.
-              See documentation in XMLOutputParser for more information.
+            parser: Parser to use for XML parsing. Can be either `'defusedxml'` or
+                `'xml'`. See documentation in `XMLOutputParser` for more information.
 
         Raises:
-            ImportError: If defusedxml is not installed and the defusedxml
-                parser is requested.
+            ImportError: If `defusedxml` is not installed and the `defusedxml` parser is
+                requested.
         """
         if parser == "defusedxml":
             if not _HAS_DEFUSEDXML:
@@ -79,10 +79,10 @@ class _StreamingParser:
         """Parse a chunk of text.
 
         Args:
-            chunk: A chunk of text to parse. This can be a string or a BaseMessage.
+            chunk: A chunk of text to parse. This can be a `str` or a `BaseMessage`.
 
         Yields:
-            A dictionary representing the parsed XML element.
+            A `dict` representing the parsed XML element.
 
         Raises:
             xml.etree.ElementTree.ParseError: If the XML is not well-formed.
@@ -147,46 +147,49 @@ class _StreamingParser:
 
 
 class XMLOutputParser(BaseTransformOutputParser):
-    """Parse an output using xml format."""
+    """Parse an output using xml format.
+
+    Returns a dictionary of tags.
+    """
 
     tags: list[str] | None = None
     """Tags to tell the LLM to expect in the XML output.
 
     Note this may not be perfect depending on the LLM implementation.
 
-    For example, with tags=["foo", "bar", "baz"]:
+    For example, with `tags=["foo", "bar", "baz"]`:
 
     1. A well-formatted XML instance:
-       "<foo>\n   <bar>\n      <baz></baz>\n   </bar>\n</foo>"
+        `"<foo>\n   <bar>\n      <baz></baz>\n   </bar>\n</foo>"`
 
     2. A badly-formatted XML instance (missing closing tag for 'bar'):
-       "<foo>\n   <bar>\n   </foo>"
+        `"<foo>\n   <bar>\n   </foo>"`
 
     3. A badly-formatted XML instance (unexpected 'tag' element):
-       "<foo>\n   <tag>\n   </tag>\n</foo>"
+        `"<foo>\n   <tag>\n   </tag>\n</foo>"`
     """
     encoding_matcher: re.Pattern = re.compile(
         r"<([^>]*encoding[^>]*)>\n(.*)", re.MULTILINE | re.DOTALL
     )
     parser: Literal["defusedxml", "xml"] = "defusedxml"
-    """Parser to use for XML parsing. Can be either 'defusedxml' or 'xml'.
+    """Parser to use for XML parsing. Can be either `'defusedxml'` or `'xml'`.
 
-    * 'defusedxml' is the default parser and is used to prevent XML vulnerabilities
-       present in some distributions of Python's standard library xml.
-       `defusedxml` is a wrapper around the standard library parser that
-       sets up the parser with secure defaults.
-    * 'xml' is the standard library parser.
+    * `'defusedxml'` is the default parser and is used to prevent XML vulnerabilities
+        present in some distributions of Python's standard library xml.
+        `defusedxml` is a wrapper around the standard library parser that
+        sets up the parser with secure defaults.
+    * `'xml'` is the standard library parser.
 
-    Use `xml` only if you are sure that your distribution of the standard library
-    is not vulnerable to XML vulnerabilities.
+    Use `xml` only if you are sure that your distribution of the standard library is not
+    vulnerable to XML vulnerabilities.
 
     Please review the following resources for more information:
 
     * https://docs.python.org/3/library/xml.html#xml-vulnerabilities
     * https://github.com/tiran/defusedxml
 
-    The standard library relies on libexpat for parsing XML:
-    https://github.com/libexpat/libexpat
+    The standard library relies on [`libexpat`](https://github.com/libexpat/libexpat)
+    for parsing XML.
     """
 
     def get_format_instructions(self) -> str:
@@ -200,12 +203,12 @@ class XMLOutputParser(BaseTransformOutputParser):
             text: The output of an LLM call.
 
         Returns:
-            A dictionary representing the parsed XML.
+            A `dict` representing the parsed XML.
 
         Raises:
             OutputParserException: If the XML is not well-formed.
-            ImportError: If defusedxml is not installed and the defusedxml
-                parser is requested.
+            ImportError: If defus`edxml is not installed and the `defusedxml` parser is
+                requested.
         """
         # Try to find XML string within triple backticks
         # Imports are temporarily placed here to avoid issue with caching on CI

libs/core/langchain_core/outputs/chat_generation.py

@@ -2,15 +2,17 @@
 
 from __future__ import annotations
 
-from typing import Literal
+from typing import TYPE_CHECKING, Literal
 
 from pydantic import model_validator
-from typing_extensions import Self
 
 from langchain_core.messages import BaseMessage, BaseMessageChunk
 from langchain_core.outputs.generation import Generation
 from langchain_core.utils._merge import merge_dicts
 
+if TYPE_CHECKING:
+    from typing_extensions import Self
+
 
 class ChatGeneration(Generation):
     """A single chat generation output.

libs/core/langchain_core/outputs/generation.py

@@ -11,9 +11,8 @@ from langchain_core.utils._merge import merge_dicts
 class Generation(Serializable):
     """A single text generation output.
 
-    Generation represents the response from an
-    `"old-fashioned" LLM <https://python.langchain.com/docs/concepts/text_llms/>__` that
-    generates regular text (not chat messages).
+    Generation represents the response from an "old-fashioned" LLM (string-in,
+    string-out) that generates regular text (not chat messages).
 
     This model is used internally by chat model and will eventually
     be mapped to a more general `LLMResult` object, and then projected into
@@ -21,8 +20,7 @@ class Generation(Serializable):
 
     LangChain users working with chat models will usually access information via
     `AIMessage` (returned from runnable interfaces) or `LLMResult` (available
-    via callbacks). Please refer the `AIMessage` and `LLMResult` schema documentation
-    for more information.
+    via callbacks). Please refer to `AIMessage` and `LLMResult` for more information.
     """
 
     text: str
@@ -35,11 +33,13 @@ class Generation(Serializable):
     """
     type: Literal["Generation"] = "Generation"
     """Type is used exclusively for serialization purposes.
-    Set to "Generation" for this class."""
+
+    Set to "Generation" for this class.
+    """
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -53,7 +53,7 @@ class Generation(Serializable):
 
 
 class GenerationChunk(Generation):
-    """Generation chunk, which can be concatenated with other Generation chunks."""
+    """`GenerationChunk`, which can be concatenated with other Generation chunks."""
 
     def __add__(self, other: GenerationChunk) -> GenerationChunk:
         """Concatenate two `GenerationChunk`s.

libs/core/langchain_core/prompt_values.py

@@ -30,15 +30,13 @@ class PromptValue(Serializable, ABC):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
     def get_lc_namespace(cls) -> list[str]:
         """Get the namespace of the LangChain object.
 
-        This is used to determine the namespace of the object when serializing.
-
         Returns:
             `["langchain", "schema", "prompt"]`
         """
@@ -50,7 +48,7 @@ class PromptValue(Serializable, ABC):
 
     @abstractmethod
     def to_messages(self) -> list[BaseMessage]:
-        """Return prompt as a list of Messages."""
+        """Return prompt as a list of messages."""
 
 
 class StringPromptValue(PromptValue):
@@ -64,8 +62,6 @@ class StringPromptValue(PromptValue):
     def get_lc_namespace(cls) -> list[str]:
         """Get the namespace of the LangChain object.
 
-        This is used to determine the namespace of the object when serializing.
-
         Returns:
             `["langchain", "prompts", "base"]`
         """
@@ -101,8 +97,6 @@ class ChatPromptValue(PromptValue):
     def get_lc_namespace(cls) -> list[str]:
         """Get the namespace of the LangChain object.
 
-        This is used to determine the namespace of the object when serializing.
-
         Returns:
             `["langchain", "prompts", "chat"]`
         """

libs/core/langchain_core/prompts/base.py

@@ -6,7 +6,7 @@ import contextlib
 import json
 import typing
 from abc import ABC, abstractmethod
-from collections.abc import Callable, Mapping
+from collections.abc import Mapping
 from functools import cached_property
 from pathlib import Path
 from typing import (
@@ -33,6 +33,8 @@ from langchain_core.runnables.config import ensure_config
 from langchain_core.utils.pydantic import create_model_v2
 
 if TYPE_CHECKING:
+    from collections.abc import Callable
+
     from langchain_core.documents import Document
 
 
@@ -46,21 +48,27 @@ class BasePromptTemplate(
 
     input_variables: list[str]
     """A list of the names of the variables whose values are required as inputs to the
-    prompt."""
+    prompt.
+    """
     optional_variables: list[str] = Field(default=[])
-    """optional_variables: A list of the names of the variables for placeholder
-       or MessagePlaceholder that are optional. These variables are auto inferred
-       from the prompt and user need not provide them."""
+    """A list of the names of the variables for placeholder or `MessagePlaceholder` that
+    are optional.
+
+    These variables are auto inferred from the prompt and user need not provide them.
+    """
     input_types: typing.Dict[str, Any] = Field(default_factory=dict, exclude=True)  # noqa: UP006
     """A dictionary of the types of the variables the prompt template expects.
-    If not provided, all variables are assumed to be strings."""
+
+    If not provided, all variables are assumed to be strings.
+    """
     output_parser: BaseOutputParser | None = None
     """How to parse the output of calling an LLM on this formatted prompt."""
     partial_variables: Mapping[str, Any] = Field(default_factory=dict)
     """A dictionary of the partial variables the prompt template carries.
 
-    Partial variables populate the template so that you don't need to
-    pass them in every time you call the prompt."""
+    Partial variables populate the template so that you don't need to pass them in every
+    time you call the prompt.
+    """
     metadata: typing.Dict[str, Any] | None = None  # noqa: UP006
     """Metadata to be used for tracing."""
     tags: list[str] | None = None
@@ -105,7 +113,7 @@ class BasePromptTemplate(
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     model_config = ConfigDict(
@@ -127,7 +135,7 @@ class BasePromptTemplate(
         """Get the input schema for the prompt.
 
         Args:
-            config: configuration for the prompt.
+            config: Configuration for the prompt.
 
         Returns:
             The input schema for the prompt.
@@ -195,8 +203,8 @@ class BasePromptTemplate(
         """Invoke the prompt.
 
         Args:
-            input: Dict, input to the prompt.
-            config: RunnableConfig, configuration for the prompt.
+            input: Input to the prompt.
+            config: Configuration for the prompt.
 
         Returns:
             The output of the prompt.
@@ -221,8 +229,8 @@ class BasePromptTemplate(
         """Async invoke the prompt.
 
         Args:
-            input: Dict, input to the prompt.
-            config: RunnableConfig, configuration for the prompt.
+            input: Input to the prompt.
+            config: Configuration for the prompt.
 
         Returns:
             The output of the prompt.
@@ -242,7 +250,7 @@ class BasePromptTemplate(
 
     @abstractmethod
     def format_prompt(self, **kwargs: Any) -> PromptValue:
-        """Create Prompt Value.
+        """Create `PromptValue`.
 
         Args:
             **kwargs: Any arguments to be passed to the prompt template.
@@ -252,7 +260,7 @@ class BasePromptTemplate(
         """
 
     async def aformat_prompt(self, **kwargs: Any) -> PromptValue:
-        """Async create Prompt Value.
+        """Async create `PromptValue`.
 
         Args:
             **kwargs: Any arguments to be passed to the prompt template.
@@ -266,7 +274,7 @@ class BasePromptTemplate(
         """Return a partial of the prompt template.
 
         Args:
-            **kwargs: partial variables to set.
+            **kwargs: Partial variables to set.
 
         Returns:
             A partial of the prompt template.
@@ -296,9 +304,9 @@ class BasePromptTemplate(
             A formatted string.
 
         Example:
-        ```python
-        prompt.format(variable1="foo")
-        ```
+            ```python
+            prompt.format(variable1="foo")
+            ```
         """
 
     async def aformat(self, **kwargs: Any) -> FormatOutputType:
@@ -311,9 +319,9 @@ class BasePromptTemplate(
             A formatted string.
 
         Example:
-        ```python
-        await prompt.aformat(variable1="foo")
-        ```
+            ```python
+            await prompt.aformat(variable1="foo")
+            ```
         """
         return self.format(**kwargs)
 
@@ -348,9 +356,9 @@ class BasePromptTemplate(
             NotImplementedError: If the prompt type is not implemented.
 
         Example:
-        ```python
-        prompt.save(file_path="path/prompt.yaml")
-        ```
+            ```python
+            prompt.save(file_path="path/prompt.yaml")
+            ```
         """
         if self.partial_variables:
             msg = "Cannot save prompt with partial variables."
@@ -402,23 +410,23 @@ def format_document(doc: Document, prompt: BasePromptTemplate[str]) -> str:
 
     First, this pulls information from the document from two sources:
 
-    1. page_content:
-        This takes the information from the `document.page_content`
-        and assigns it to a variable named `page_content`.
-    2. metadata:
-        This takes information from `document.metadata` and assigns
-        it to variables of the same name.
+    1. `page_content`:
+        This takes the information from the `document.page_content` and assigns it to a
+        variable named `page_content`.
+    2. `metadata`:
+        This takes information from `document.metadata` and assigns it to variables of
+        the same name.
 
     Those variables are then passed into the `prompt` to produce a formatted string.
 
     Args:
-        doc: Document, the page_content and metadata will be used to create
+        doc: `Document`, the `page_content` and `metadata` will be used to create
             the final string.
-        prompt: BasePromptTemplate, will be used to format the page_content
-            and metadata into the final string.
+        prompt: `BasePromptTemplate`, will be used to format the `page_content`
+            and `metadata` into the final string.
 
     Returns:
-        string of the document formatted.
+        String of the document formatted.
 
     Example:
         ```python
@@ -429,7 +437,6 @@ def format_document(doc: Document, prompt: BasePromptTemplate[str]) -> str:
         prompt = PromptTemplate.from_template("Page {page}: {page_content}")
         format_document(doc, prompt)
         >>> "Page 1: This is a joke"
-
         ```
     """
     return prompt.format(**_get_document_info(doc, prompt))
@@ -440,22 +447,22 @@ async def aformat_document(doc: Document, prompt: BasePromptTemplate[str]) -> st
 
     First, this pulls information from the document from two sources:
 
-    1. page_content:
-        This takes the information from the `document.page_content`
-        and assigns it to a variable named `page_content`.
-    2. metadata:
-        This takes information from `document.metadata` and assigns
-        it to variables of the same name.
+    1. `page_content`:
+        This takes the information from the `document.page_content` and assigns it to a
+        variable named `page_content`.
+    2. `metadata`:
+        This takes information from `document.metadata` and assigns it to variables of
+        the same name.
 
     Those variables are then passed into the `prompt` to produce a formatted string.
 
     Args:
-        doc: Document, the page_content and metadata will be used to create
+        doc: `Document`, the `page_content` and `metadata` will be used to create
             the final string.
-        prompt: BasePromptTemplate, will be used to format the page_content
-            and metadata into the final string.
+        prompt: `BasePromptTemplate`, will be used to format the `page_content`
+            and `metadata` into the final string.
 
     Returns:
-        string of the document formatted.
+        String of the document formatted.
     """
     return await prompt.aformat(**_get_document_info(doc, prompt))

libs/core/langchain_core/prompts/chat.py

@@ -587,14 +587,15 @@ class _StringImageMessagePromptTemplate(BaseMessagePromptTemplate):
         for prompt in self.prompt:
             inputs = {var: kwargs[var] for var in prompt.input_variables}
             if isinstance(prompt, StringPromptTemplate):
-                formatted: str | ImageURL | dict[str, Any] = prompt.format(**inputs)
-                content.append({"type": "text", "text": formatted})
+                formatted_text: str = prompt.format(**inputs)
+                if formatted_text != "":
+                    content.append({"type": "text", "text": formatted_text})
             elif isinstance(prompt, ImagePromptTemplate):
-                formatted = prompt.format(**inputs)
-                content.append({"type": "image_url", "image_url": formatted})
+                formatted_image: ImageURL = prompt.format(**inputs)
+                content.append({"type": "image_url", "image_url": formatted_image})
             elif isinstance(prompt, DictPromptTemplate):
-                formatted = prompt.format(**inputs)
-                content.append(formatted)
+                formatted_dict: dict[str, Any] = prompt.format(**inputs)
+                content.append(formatted_dict)
         return self._msg_class(
             content=content, additional_kwargs=self.additional_kwargs
         )
@@ -617,16 +618,15 @@ class _StringImageMessagePromptTemplate(BaseMessagePromptTemplate):
         for prompt in self.prompt:
             inputs = {var: kwargs[var] for var in prompt.input_variables}
             if isinstance(prompt, StringPromptTemplate):
-                formatted: str | ImageURL | dict[str, Any] = await prompt.aformat(
-                    **inputs
-                )
-                content.append({"type": "text", "text": formatted})
+                formatted_text: str = await prompt.aformat(**inputs)
+                if formatted_text != "":
+                    content.append({"type": "text", "text": formatted_text})
             elif isinstance(prompt, ImagePromptTemplate):
-                formatted = await prompt.aformat(**inputs)
-                content.append({"type": "image_url", "image_url": formatted})
+                formatted_image: ImageURL = await prompt.aformat(**inputs)
+                content.append({"type": "image_url", "image_url": formatted_image})
             elif isinstance(prompt, DictPromptTemplate):
-                formatted = prompt.format(**inputs)
-                content.append(formatted)
+                formatted_dict: dict[str, Any] = prompt.format(**inputs)
+                content.append(formatted_dict)
         return self._msg_class(
             content=content, additional_kwargs=self.additional_kwargs
         )
@@ -776,42 +776,36 @@ class ChatPromptTemplate(BaseChatPromptTemplate):
 
     Use to create flexible templated prompts for chat models.
 
-    Examples:
-        !!! warning "Behavior changed in 0.2.24"
-            You can pass any Message-like formats supported by
-            `ChatPromptTemplate.from_messages()` directly to `ChatPromptTemplate()`
-            init.
+    ```python
+    from langchain_core.prompts import ChatPromptTemplate
 
-        ```python
-        from langchain_core.prompts import ChatPromptTemplate
+    template = ChatPromptTemplate(
+        [
+            ("system", "You are a helpful AI bot. Your name is {name}."),
+            ("human", "Hello, how are you doing?"),
+            ("ai", "I'm doing well, thanks!"),
+            ("human", "{user_input}"),
+        ]
+    )
 
-        template = ChatPromptTemplate(
-            [
-                ("system", "You are a helpful AI bot. Your name is {name}."),
-                ("human", "Hello, how are you doing?"),
-                ("ai", "I'm doing well, thanks!"),
-                ("human", "{user_input}"),
-            ]
-        )
+    prompt_value = template.invoke(
+        {
+            "name": "Bob",
+            "user_input": "What is your name?",
+        }
+    )
+    # Output:
+    # ChatPromptValue(
+    #    messages=[
+    #        SystemMessage(content='You are a helpful AI bot. Your name is Bob.'),
+    #        HumanMessage(content='Hello, how are you doing?'),
+    #        AIMessage(content="I'm doing well, thanks!"),
+    #        HumanMessage(content='What is your name?')
+    #    ]
+    # )
+    ```
 
-        prompt_value = template.invoke(
-            {
-                "name": "Bob",
-                "user_input": "What is your name?",
-            }
-        )
-        # Output:
-        # ChatPromptValue(
-        #    messages=[
-        #        SystemMessage(content='You are a helpful AI bot. Your name is Bob.'),
-        #        HumanMessage(content='Hello, how are you doing?'),
-        #        AIMessage(content="I'm doing well, thanks!"),
-        #        HumanMessage(content='What is your name?')
-        #    ]
-        # )
-        ```
-
-    Messages Placeholder:
+    !!! note "Messages Placeholder"
 
         ```python
         # In addition to Human/AI/Tool/Function messages,
@@ -852,13 +846,12 @@ class ChatPromptTemplate(BaseChatPromptTemplate):
         # )
         ```
 
-    Single-variable template:
+    !!! note "Single-variable template"
 
         If your prompt has only a single input variable (i.e., 1 instance of "{variable_nams}"),
         and you invoke the template with a non-dict object, the prompt template will
         inject the provided argument into that variable location.
 
-
         ```python
         from langchain_core.prompts import ChatPromptTemplate
 
@@ -898,25 +891,40 @@ class ChatPromptTemplate(BaseChatPromptTemplate):
         """Create a chat prompt template from a variety of message formats.
 
         Args:
-            messages: sequence of message representations.
+            messages: Sequence of message representations.
+
                 A message can be represented using the following formats:
-                (1) BaseMessagePromptTemplate, (2) BaseMessage, (3) 2-tuple of
-                (message type, template); e.g., ("human", "{user_input}"),
-                (4) 2-tuple of (message class, template), (5) a string which is
-                shorthand for ("human", template); e.g., "{user_input}".
-            template_format: format of the template.
-            input_variables: A list of the names of the variables whose values are
-                required as inputs to the prompt.
-            optional_variables: A list of the names of the variables for placeholder
-                or MessagePlaceholder that are optional.
-                These variables are auto inferred from the prompt and user need not
-                provide them.
-            partial_variables: A dictionary of the partial variables the prompt
-                template carries. Partial variables populate the template so that you
-                don't need to pass them in every time you call the prompt.
-            validate_template: Whether to validate the template.
-            input_types: A dictionary of the types of the variables the prompt template
-                expects. If not provided, all variables are assumed to be strings.
+
+                1. `BaseMessagePromptTemplate`
+                2. `BaseMessage`
+                3. 2-tuple of `(message type, template)`; e.g.,
+                    `("human", "{user_input}")`
+                4. 2-tuple of `(message class, template)`
+                5. A string which is shorthand for `("human", template)`; e.g.,
+                    `"{user_input}"`
+            template_format: Format of the template.
+            **kwargs: Additional keyword arguments passed to `BasePromptTemplate`,
+                including (but not limited to):
+
+                - `input_variables`: A list of the names of the variables whose values
+                    are required as inputs to the prompt.
+                - `optional_variables`: A list of the names of the variables for
+                    placeholder or `MessagePlaceholder` that are optional.
+
+                    These variables are auto inferred from the prompt and user need not
+                    provide them.
+
+                - `partial_variables`: A dictionary of the partial variables the prompt
+                    template carries.
+
+                    Partial variables populate the template so that you don't need to
+                    pass them in every time you call the prompt.
+
+                - `validate_template`: Whether to validate the template.
+                - `input_types`: A dictionary of the types of the variables the prompt
+                    template expects.
+
+                    If not provided, all variables are assumed to be strings.
 
         Examples:
             Instantiation from a list of message templates:
@@ -1121,12 +1129,17 @@ class ChatPromptTemplate(BaseChatPromptTemplate):
             )
             ```
         Args:
-            messages: sequence of message representations.
+            messages: Sequence of message representations.
+
                 A message can be represented using the following formats:
-                (1) BaseMessagePromptTemplate, (2) BaseMessage, (3) 2-tuple of
-                (message type, template); e.g., ("human", "{user_input}"),
-                (4) 2-tuple of (message class, template), (5) a string which is
-                shorthand for ("human", template); e.g., "{user_input}".
+
+                1. `BaseMessagePromptTemplate`
+                2. `BaseMessage`
+                3. 2-tuple of `(message type, template)`; e.g.,
+                    `("human", "{user_input}")`
+                4. 2-tuple of `(message class, template)`
+                5. A string which is shorthand for `("human", template)`; e.g.,
+                    `"{user_input}"`
             template_format: format of the template.
 
         Returns:
@@ -1238,7 +1251,7 @@ class ChatPromptTemplate(BaseChatPromptTemplate):
         """Extend the chat template with a sequence of messages.
 
         Args:
-            messages: sequence of message representations to append.
+            messages: Sequence of message representations to append.
         """
         self.messages.extend(
             [_convert_to_message_template(message) for message in messages]
@@ -1335,11 +1348,25 @@ def _create_template_from_message_type(
                 raise ValueError(msg)
             var_name = template[1:-1]
             message = MessagesPlaceholder(variable_name=var_name, optional=True)
-        elif len(template) == 2 and isinstance(template[1], bool):
-            var_name_wrapped, is_optional = template
+        else:
+            try:
+                var_name_wrapped, is_optional = template
+            except ValueError as e:
+                msg = (
+                    "Unexpected arguments for placeholder message type."
+                    " Expected either a single string variable name"
+                    " or a list of [variable_name: str, is_optional: bool]."
+                    f" Got: {template}"
+                )
+                raise ValueError(msg) from e
+
+            if not isinstance(is_optional, bool):
+                msg = f"Expected is_optional to be a boolean. Got: {is_optional}"
+                raise ValueError(msg)  # noqa: TRY004
+
             if not isinstance(var_name_wrapped, str):
                 msg = f"Expected variable name to be a string. Got: {var_name_wrapped}"
-                raise ValueError(msg)  # noqa:TRY004
+                raise ValueError(msg)  # noqa: TRY004
             if var_name_wrapped[0] != "{" or var_name_wrapped[-1] != "}":
                 msg = (
                     f"Invalid placeholder template: {var_name_wrapped}."
@@ -1349,14 +1376,6 @@ def _create_template_from_message_type(
             var_name = var_name_wrapped[1:-1]
 
             message = MessagesPlaceholder(variable_name=var_name, optional=is_optional)
-        else:
-            msg = (
-                "Unexpected arguments for placeholder message type."
-                " Expected either a single string variable name"
-                " or a list of [variable_name: str, is_optional: bool]."
-                f" Got: {template}"
-            )
-            raise ValueError(msg)
     else:
         msg = (
             f"Unexpected message type: {message_type}. Use one of 'human',"
@@ -1410,10 +1429,11 @@ def _convert_to_message_template(
                 )
                 raise ValueError(msg)
             message = (message["role"], message["content"])
-        if len(message) != 2:
+        try:
+            message_type_str, template = message
+        except ValueError as e:
             msg = f"Expected 2-tuple of (role, template), got {message}"
-            raise ValueError(msg)
-        message_type_str, template = message
+            raise ValueError(msg) from e
         if isinstance(message_type_str, str):
             message_ = _create_template_from_message_type(
                 message_type_str, template, template_format=template_format

libs/core/langchain_core/prompts/dict.py

@@ -69,7 +69,7 @@ class DictPromptTemplate(RunnableSerializable[dict, dict]):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod

libs/core/langchain_core/prompts/message.py

@@ -6,10 +6,10 @@ from abc import ABC, abstractmethod
 from typing import TYPE_CHECKING, Any
 
 from langchain_core.load import Serializable
-from langchain_core.messages import BaseMessage
 from langchain_core.utils.interactive_env import is_interactive_env
 
 if TYPE_CHECKING:
+    from langchain_core.messages import BaseMessage
     from langchain_core.prompts.chat import ChatPromptTemplate
 
 
@@ -18,7 +18,7 @@ class BaseMessagePromptTemplate(Serializable, ABC):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -32,13 +32,13 @@ class BaseMessagePromptTemplate(Serializable, ABC):
 
     @abstractmethod
     def format_messages(self, **kwargs: Any) -> list[BaseMessage]:
-        """Format messages from kwargs. Should return a list of BaseMessages.
+        """Format messages from kwargs. Should return a list of `BaseMessage` objects.
 
         Args:
             **kwargs: Keyword arguments to use for formatting.
 
         Returns:
-            List of BaseMessages.
+            List of `BaseMessage` objects.
         """
 
     async def aformat_messages(self, **kwargs: Any) -> list[BaseMessage]:
@@ -48,7 +48,7 @@ class BaseMessagePromptTemplate(Serializable, ABC):
             **kwargs: Keyword arguments to use for formatting.
 
         Returns:
-            List of BaseMessages.
+            List of `BaseMessage` objects.
         """
         return self.format_messages(**kwargs)
 

libs/core/langchain_core/prompts/string.py

@@ -4,9 +4,8 @@ from __future__ import annotations
 
 import warnings
 from abc import ABC
-from collections.abc import Callable, Sequence
 from string import Formatter
-from typing import Any, Literal
+from typing import TYPE_CHECKING, Any, Literal
 
 from pydantic import BaseModel, create_model
 
@@ -16,10 +15,70 @@ from langchain_core.utils import get_colored_text, mustache
 from langchain_core.utils.formatting import formatter
 from langchain_core.utils.interactive_env import is_interactive_env
 
+if TYPE_CHECKING:
+    from collections.abc import Callable, Sequence
+
 try:
-    from jinja2 import Environment, meta
+    from jinja2 import meta
+    from jinja2.exceptions import SecurityError
     from jinja2.sandbox import SandboxedEnvironment
 
+    class _RestrictedSandboxedEnvironment(SandboxedEnvironment):
+        """A more restrictive Jinja2 sandbox that blocks all attribute/method access.
+
+        This sandbox only allows simple variable lookups, no attribute or method access.
+        This prevents template injection attacks via methods like parse_raw().
+        """
+
+        def is_safe_attribute(self, _obj: Any, _attr: str, _value: Any) -> bool:
+            """Block ALL attribute access for security.
+
+            Only allow accessing variables directly from the context dict,
+            no attribute access on those objects.
+
+            Args:
+                _obj: The object being accessed (unused, always blocked).
+                _attr: The attribute name (unused, always blocked).
+                _value: The attribute value (unused, always blocked).
+
+            Returns:
+                False - all attribute access is blocked.
+            """
+            # Block all attribute access
+            return False
+
+        def is_safe_callable(self, _obj: Any) -> bool:
+            """Block all method calls for security.
+
+            Args:
+                _obj: The object being checked (unused, always blocked).
+
+            Returns:
+                False - all callables are blocked.
+            """
+            return False
+
+        def getattr(self, obj: Any, attribute: str) -> Any:
+            """Override getattr to block all attribute access.
+
+            Args:
+                obj: The object.
+                attribute: The attribute name.
+
+            Returns:
+                Never returns.
+
+            Raises:
+                SecurityError: Always, to block attribute access.
+            """
+            msg = (
+                f"Access to attributes is not allowed in templates. "
+                f"Attempted to access '{attribute}' on {type(obj).__name__}. "
+                f"Use only simple variable names like {{{{variable}}}} "
+                f"without dots or methods."
+            )
+            raise SecurityError(msg)
+
     _HAS_JINJA2 = True
 except ImportError:
     _HAS_JINJA2 = False
@@ -59,14 +118,10 @@ def jinja2_formatter(template: str, /, **kwargs: Any) -> str:
         )
         raise ImportError(msg)
 
-    # This uses a sandboxed environment to prevent arbitrary code execution.
-    # Jinja2 uses an opt-out rather than opt-in approach for sand-boxing.
-    # Please treat this sand-boxing as a best-effort approach rather than
-    # a guarantee of security.
-    # We recommend to never use jinja2 templates with untrusted inputs.
-    # https://jinja.palletsprojects.com/en/3.1.x/sandbox/
-    # approach not a guarantee of security.
-    return SandboxedEnvironment().from_string(template).render(**kwargs)
+    # Use a restricted sandbox that blocks ALL attribute/method access
+    # Only simple variable lookups like {{variable}} are allowed
+    # Attribute access like {{variable.attr}} or {{variable.method()}} is blocked
+    return _RestrictedSandboxedEnvironment().from_string(template).render(**kwargs)
 
 
 def validate_jinja2(template: str, input_variables: list[str]) -> None:
@@ -101,7 +156,7 @@ def _get_jinja2_variables_from_template(template: str) -> set[str]:
             "Please install it with `pip install jinja2`."
         )
         raise ImportError(msg)
-    env = Environment()  # noqa: S701
+    env = _RestrictedSandboxedEnvironment()
     ast = env.parse(template)
     return meta.find_undeclared_variables(ast)
 
@@ -122,13 +177,16 @@ def mustache_formatter(template: str, /, **kwargs: Any) -> str:
 def mustache_template_vars(
     template: str,
 ) -> set[str]:
-    """Get the variables from a mustache template.
+    """Get the top-level variables from a mustache template.
+
+    For nested variables like `{{person.name}}`, only the top-level
+    key (`person`) is returned.
 
     Args:
         template: The template string.
 
     Returns:
-        The variables from the template.
+       The top-level variables from the template.
     """
     variables: set[str] = set()
     section_depth = 0
@@ -268,6 +326,30 @@ def get_template_variables(template: str, template_format: str) -> list[str]:
         msg = f"Unsupported template format: {template_format}"
         raise ValueError(msg)
 
+    # For f-strings, block attribute access and indexing syntax
+    # This prevents template injection attacks via accessing dangerous attributes
+    if template_format == "f-string":
+        for var in input_variables:
+            # Formatter().parse() returns field names with dots/brackets if present
+            # e.g., "obj.attr" or "obj[0]" - we need to block these
+            if "." in var or "[" in var or "]" in var:
+                msg = (
+                    f"Invalid variable name {var!r} in f-string template. "
+                    f"Variable names cannot contain attribute "
+                    f"access (.) or indexing ([])."
+                )
+                raise ValueError(msg)
+
+            # Block variable names that are all digits (e.g., "0", "100")
+            # These are interpreted as positional arguments, not keyword arguments
+            if var.isdigit():
+                msg = (
+                    f"Invalid variable name {var!r} in f-string template. "
+                    f"Variable names cannot be all digits as they are interpreted "
+                    f"as positional arguments."
+                )
+                raise ValueError(msg)
+
     return sorted(input_variables)
 
 

libs/core/langchain_core/prompts/structured.py

@@ -49,7 +49,13 @@ class StructuredPrompt(ChatPromptTemplate):
             structured_output_kwargs: additional kwargs for structured output.
             template_format: template format for the prompt.
         """
-        schema_ = schema_ or kwargs.pop("schema")
+        schema_ = schema_ or kwargs.pop("schema", None)
+        if not schema_:
+            err_msg = (
+                "Must pass in a non-empty structured output schema. Received: "
+                f"{schema_}"
+            )
+            raise ValueError(err_msg)
         structured_output_kwargs = structured_output_kwargs or {}
         for k in set(kwargs).difference(get_pydantic_field_names(self.__class__)):
             structured_output_kwargs[k] = kwargs.pop(k)
@@ -104,19 +110,23 @@ class StructuredPrompt(ChatPromptTemplate):
             )
             ```
         Args:
-            messages: sequence of message representations.
+            messages: Sequence of message representations.
+
                 A message can be represented using the following formats:
-                (1) BaseMessagePromptTemplate, (2) BaseMessage, (3) 2-tuple of
-                (message type, template); e.g., ("human", "{user_input}"),
-                (4) 2-tuple of (message class, template), (5) a string which is
-                shorthand for ("human", template); e.g., "{user_input}"
-            schema: a dictionary representation of function call, or a Pydantic model.
+
+                1. `BaseMessagePromptTemplate`
+                2. `BaseMessage`
+                3. 2-tuple of `(message type, template)`; e.g.,
+                    `("human", "{user_input}")`
+                4. 2-tuple of `(message class, template)`
+                5. A string which is shorthand for `("human", template)`; e.g.,
+                    `"{user_input}"`
+            schema: A dictionary representation of function call, or a Pydantic model.
             **kwargs: Any additional kwargs to pass through to
                 `ChatModel.with_structured_output(schema, **kwargs)`.
 
         Returns:
-            a structured prompt template
-
+            A structured prompt template
         """
         return cls(messages, schema, **kwargs)
 

libs/core/langchain_core/rate_limiters.py

@@ -105,7 +105,9 @@ class InMemoryRateLimiter(BaseRateLimiter):
 
         from langchain_anthropic import ChatAnthropic
 
-        model = ChatAnthropic(model_name="claude-sonnet-4-5", rate_limiter=rate_limiter)
+        model = ChatAnthropic(
+            model_name="claude-sonnet-4-5-20250929", rate_limiter=rate_limiter
+        )
 
         for _ in range(5):
             tic = time.time()

libs/core/langchain_core/retrievers.py

@@ -50,65 +50,65 @@ class LangSmithRetrieverParams(TypedDict, total=False):
 
 
 class BaseRetriever(RunnableSerializable[RetrieverInput, RetrieverOutput], ABC):
-    """Abstract base class for a Document retrieval system.
+    """Abstract base class for a document retrieval system.
 
     A retrieval system is defined as something that can take string queries and return
-    the most 'relevant' Documents from some source.
+    the most 'relevant' documents from some source.
 
     Usage:
 
-    A retriever follows the standard Runnable interface, and should be used
-    via the standard Runnable methods of `invoke`, `ainvoke`, `batch`, `abatch`.
+    A retriever follows the standard `Runnable` interface, and should be used via the
+    standard `Runnable` methods of `invoke`, `ainvoke`, `batch`, `abatch`.
 
     Implementation:
 
-    When implementing a custom retriever, the class should implement
-    the `_get_relevant_documents` method to define the logic for retrieving documents.
+    When implementing a custom retriever, the class should implement the
+    `_get_relevant_documents` method to define the logic for retrieving documents.
 
     Optionally, an async native implementations can be provided by overriding the
     `_aget_relevant_documents` method.
 
-    Example: A retriever that returns the first 5 documents from a list of documents
+    !!! example "Retriever that returns the first 5 documents from a list of documents"
 
-    ```python
-    from langchain_core.documents import Document
-    from langchain_core.retrievers import BaseRetriever
+        ```python
+        from langchain_core.documents import Document
+        from langchain_core.retrievers import BaseRetriever
 
-    class SimpleRetriever(BaseRetriever):
-        docs: list[Document]
-        k: int = 5
+        class SimpleRetriever(BaseRetriever):
+            docs: list[Document]
+            k: int = 5
 
-        def _get_relevant_documents(self, query: str) -> list[Document]:
-            \"\"\"Return the first k documents from the list of documents\"\"\"
-            return self.docs[:self.k]
+            def _get_relevant_documents(self, query: str) -> list[Document]:
+                \"\"\"Return the first k documents from the list of documents\"\"\"
+                return self.docs[:self.k]
 
-        async def _aget_relevant_documents(self, query: str) -> list[Document]:
-            \"\"\"(Optional) async native implementation.\"\"\"
-            return self.docs[:self.k]
-    ```
+            async def _aget_relevant_documents(self, query: str) -> list[Document]:
+                \"\"\"(Optional) async native implementation.\"\"\"
+                return self.docs[:self.k]
+        ```
 
-    Example: A simple retriever based on a scikit-learn vectorizer
+    !!! example "Simple retriever based on a scikit-learn vectorizer"
 
-    ```python
-    from sklearn.metrics.pairwise import cosine_similarity
+        ```python
+        from sklearn.metrics.pairwise import cosine_similarity
 
 
-    class TFIDFRetriever(BaseRetriever, BaseModel):
-        vectorizer: Any
-        docs: list[Document]
-        tfidf_array: Any
-        k: int = 4
+        class TFIDFRetriever(BaseRetriever, BaseModel):
+            vectorizer: Any
+            docs: list[Document]
+            tfidf_array: Any
+            k: int = 4
 
-        class Config:
-            arbitrary_types_allowed = True
+            class Config:
+                arbitrary_types_allowed = True
 
-        def _get_relevant_documents(self, query: str) -> list[Document]:
-            # Ip -- (n_docs,x), Op -- (n_docs,n_Feats)
-            query_vec = self.vectorizer.transform([query])
-            # Op -- (n_docs,1) -- Cosine Sim with each doc
-            results = cosine_similarity(self.tfidf_array, query_vec).reshape((-1,))
-            return [self.docs[i] for i in results.argsort()[-self.k :][::-1]]
-    ```
+            def _get_relevant_documents(self, query: str) -> list[Document]:
+                # Ip -- (n_docs,x), Op -- (n_docs,n_Feats)
+                query_vec = self.vectorizer.transform([query])
+                # Op -- (n_docs,1) -- Cosine Sim with each doc
+                results = cosine_similarity(self.tfidf_array, query_vec).reshape((-1,))
+                return [self.docs[i] for i in results.argsort()[-self.k :][::-1]]
+        ```
     """
 
     model_config = ConfigDict(
@@ -119,15 +119,19 @@ class BaseRetriever(RunnableSerializable[RetrieverInput, RetrieverOutput], ABC):
     _expects_other_args: bool = False
     tags: list[str] | None = None
     """Optional list of tags associated with the retriever.
+
     These tags will be associated with each call to this retriever,
     and passed as arguments to the handlers defined in `callbacks`.
+
     You can use these to eg identify a specific instance of a retriever with its
     use case.
     """
     metadata: dict[str, Any] | None = None
     """Optional metadata associated with the retriever.
+
     This metadata will be associated with each call to this retriever,
     and passed as arguments to the handlers defined in `callbacks`.
+
     You can use these to eg identify a specific instance of a retriever with its
     use case.
     """

libs/core/langchain_core/runnables/base.py

@@ -118,6 +118,8 @@ if TYPE_CHECKING:
 
 Other = TypeVar("Other")
 
+_RUNNABLE_GENERIC_NUM_ARGS = 2  # Input and Output
+
 
 class Runnable(ABC, Generic[Input, Output]):
     """A unit of work that can be invoked, batched, streamed, transformed and composed.
@@ -147,11 +149,11 @@ class Runnable(ABC, Generic[Input, Output]):
     the `input_schema` property, the `output_schema` property and `config_schema`
     method.
 
-    LCEL and Composition
-    ====================
+    Composition
+    ===========
+
+    Runnable objects can be composed together to create chains in a declarative way.
 
-    The LangChain Expression Language (LCEL) is a declarative way to compose
-    `Runnable` objectsinto chains.
     Any chain constructed this way will automatically have sync, async, batch, and
     streaming support.
 
@@ -235,21 +237,21 @@ class Runnable(ABC, Generic[Input, Output]):
 
     You can set the global debug flag to True to enable debug output for all chains:
 
-        ```python
-        from langchain_core.globals import set_debug
+    ```python
+    from langchain_core.globals import set_debug
 
-        set_debug(True)
-        ```
+    set_debug(True)
+    ```
 
     Alternatively, you can pass existing or custom callbacks to any given chain:
 
-        ```python
-        from langchain_core.tracers import ConsoleCallbackHandler
+    ```python
+    from langchain_core.tracers import ConsoleCallbackHandler
 
-        chain.invoke(..., config={"callbacks": [ConsoleCallbackHandler()]})
-        ```
+    chain.invoke(..., config={"callbacks": [ConsoleCallbackHandler()]})
+    ```
 
-    For a UI (and much more) checkout [LangSmith](https://docs.smith.langchain.com/).
+    For a UI (and much more) checkout [LangSmith](https://docs.langchain.com/langsmith/home).
 
     """
 
@@ -309,7 +311,10 @@ class Runnable(ABC, Generic[Input, Output]):
         for base in self.__class__.mro():
             if hasattr(base, "__pydantic_generic_metadata__"):
                 metadata = base.__pydantic_generic_metadata__
-                if "args" in metadata and len(metadata["args"]) == 2:
+                if (
+                    "args" in metadata
+                    and len(metadata["args"]) == _RUNNABLE_GENERIC_NUM_ARGS
+                ):
                     return metadata["args"][0]
 
         # If we didn't find a Pydantic model in the parent classes,
@@ -317,7 +322,7 @@ class Runnable(ABC, Generic[Input, Output]):
         # Runnables that are not pydantic models.
         for cls in self.__class__.__orig_bases__:  # type: ignore[attr-defined]
             type_args = get_args(cls)
-            if type_args and len(type_args) == 2:
+            if type_args and len(type_args) == _RUNNABLE_GENERIC_NUM_ARGS:
                 return type_args[0]
 
         msg = (
@@ -340,12 +345,15 @@ class Runnable(ABC, Generic[Input, Output]):
         for base in self.__class__.mro():
             if hasattr(base, "__pydantic_generic_metadata__"):
                 metadata = base.__pydantic_generic_metadata__
-                if "args" in metadata and len(metadata["args"]) == 2:
+                if (
+                    "args" in metadata
+                    and len(metadata["args"]) == _RUNNABLE_GENERIC_NUM_ARGS
+                ):
                     return metadata["args"][1]
 
         for cls in self.__class__.__orig_bases__:  # type: ignore[attr-defined]
             type_args = get_args(cls)
-            if type_args and len(type_args) == 2:
+            if type_args and len(type_args) == _RUNNABLE_GENERIC_NUM_ARGS:
                 return type_args[1]
 
         msg = (
@@ -424,7 +432,7 @@ class Runnable(ABC, Generic[Input, Output]):
             print(runnable.get_input_jsonschema())
             ```
 
-        !!! version-added "Added in version 0.3.0"
+        !!! version-added "Added in `langchain-core` 0.3.0"
 
         """
         return self.get_input_schema(config).model_json_schema()
@@ -502,7 +510,7 @@ class Runnable(ABC, Generic[Input, Output]):
             print(runnable.get_output_jsonschema())
             ```
 
-        !!! version-added "Added in version 0.3.0"
+        !!! version-added "Added in `langchain-core` 0.3.0"
 
         """
         return self.get_output_schema(config).model_json_schema()
@@ -566,7 +574,7 @@ class Runnable(ABC, Generic[Input, Output]):
         Returns:
             A JSON schema that represents the config of the `Runnable`.
 
-        !!! version-added "Added in version 0.3.0"
+        !!! version-added "Added in `langchain-core` 0.3.0"
 
         """
         return self.config_schema(include=include).model_json_schema()
@@ -699,51 +707,53 @@ class Runnable(ABC, Generic[Input, Output]):
     def pick(self, keys: str | list[str]) -> RunnableSerializable[Any, Any]:
         """Pick keys from the output `dict` of this `Runnable`.
 
-        Pick a single key:
+        !!! example "Pick a single key"
 
-        ```python
-        import json
+            ```python
+            import json
 
-        from langchain_core.runnables import RunnableLambda, RunnableMap
+            from langchain_core.runnables import RunnableLambda, RunnableMap
 
-        as_str = RunnableLambda(str)
-        as_json = RunnableLambda(json.loads)
-        chain = RunnableMap(str=as_str, json=as_json)
+            as_str = RunnableLambda(str)
+            as_json = RunnableLambda(json.loads)
+            chain = RunnableMap(str=as_str, json=as_json)
 
-        chain.invoke("[1, 2, 3]")
-        # -> {"str": "[1, 2, 3]", "json": [1, 2, 3]}
+            chain.invoke("[1, 2, 3]")
+            # -> {"str": "[1, 2, 3]", "json": [1, 2, 3]}
 
-        json_only_chain = chain.pick("json")
-        json_only_chain.invoke("[1, 2, 3]")
-        # -> [1, 2, 3]
-        ```
+            json_only_chain = chain.pick("json")
+            json_only_chain.invoke("[1, 2, 3]")
+            # -> [1, 2, 3]
+            ```
 
-        Pick a list of keys:
+        !!! example "Pick a list of keys"
 
-        ```python
-        from typing import Any
+            ```python
+            from typing import Any
 
-        import json
+            import json
 
-        from langchain_core.runnables import RunnableLambda, RunnableMap
+            from langchain_core.runnables import RunnableLambda, RunnableMap
 
-        as_str = RunnableLambda(str)
-        as_json = RunnableLambda(json.loads)
+            as_str = RunnableLambda(str)
+            as_json = RunnableLambda(json.loads)
 
 
-        def as_bytes(x: Any) -> bytes:
-            return bytes(x, "utf-8")
+            def as_bytes(x: Any) -> bytes:
+                return bytes(x, "utf-8")
 
 
-        chain = RunnableMap(str=as_str, json=as_json, bytes=RunnableLambda(as_bytes))
+            chain = RunnableMap(
+                str=as_str, json=as_json, bytes=RunnableLambda(as_bytes)
+            )
 
-        chain.invoke("[1, 2, 3]")
-        # -> {"str": "[1, 2, 3]", "json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
+            chain.invoke("[1, 2, 3]")
+            # -> {"str": "[1, 2, 3]", "json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
 
-        json_and_bytes_chain = chain.pick(["json", "bytes"])
-        json_and_bytes_chain.invoke("[1, 2, 3]")
-        # -> {"json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
-        ```
+            json_and_bytes_chain = chain.pick(["json", "bytes"])
+            json_and_bytes_chain.invoke("[1, 2, 3]")
+            # -> {"json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
+            ```
 
         Args:
             keys: A key or list of keys to pick from the output dict.
@@ -766,7 +776,7 @@ class Runnable(ABC, Generic[Input, Output]):
         """Assigns new fields to the `dict` output of this `Runnable`.
 
         ```python
-        from langchain_community.llms.fake import FakeStreamingListLLM
+        from langchain_core.language_models.fake import FakeStreamingListLLM
         from langchain_core.output_parsers import StrOutputParser
         from langchain_core.prompts import SystemMessagePromptTemplate
         from langchain_core.runnables import Runnable
@@ -818,10 +828,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             input: The input to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
 
         Returns:
             The output of the `Runnable`.
@@ -838,10 +850,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             input: The input to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
 
         Returns:
             The output of the `Runnable`.
@@ -868,8 +882,9 @@ class Runnable(ABC, Generic[Input, Output]):
             config: A config to use when invoking the `Runnable`. The config supports
                 standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work
-                to do in parallel, and other keys. Please refer to the
-                `RunnableConfig` for more details.
+                to do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -932,10 +947,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             inputs: A list of inputs to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -998,10 +1015,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             inputs: A list of inputs to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -1061,10 +1080,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             inputs: A list of inputs to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -1353,48 +1374,50 @@ class Runnable(ABC, Generic[Input, Output]):
         ).with_config({"run_name": "my_template", "tags": ["my_template"]})
         ```
 
-        For instance:
+        !!! example
 
-        ```python
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from langchain_core.runnables import RunnableLambda
 
 
-        async def reverse(s: str) -> str:
-            return s[::-1]
+            async def reverse(s: str) -> str:
+                return s[::-1]
 
 
-        chain = RunnableLambda(func=reverse)
+            chain = RunnableLambda(func=reverse)
 
-        events = [event async for event in chain.astream_events("hello", version="v2")]
+            events = [
+                event async for event in chain.astream_events("hello", version="v2")
+            ]
 
-        # Will produce the following events
-        # (run_id, and parent_ids has been omitted for brevity):
-        [
-            {
-                "data": {"input": "hello"},
-                "event": "on_chain_start",
-                "metadata": {},
-                "name": "reverse",
-                "tags": [],
-            },
-            {
-                "data": {"chunk": "olleh"},
-                "event": "on_chain_stream",
-                "metadata": {},
-                "name": "reverse",
-                "tags": [],
-            },
-            {
-                "data": {"output": "olleh"},
-                "event": "on_chain_end",
-                "metadata": {},
-                "name": "reverse",
-                "tags": [],
-            },
-        ]
-        ```
+            # Will produce the following events
+            # (run_id, and parent_ids has been omitted for brevity):
+            [
+                {
+                    "data": {"input": "hello"},
+                    "event": "on_chain_start",
+                    "metadata": {},
+                    "name": "reverse",
+                    "tags": [],
+                },
+                {
+                    "data": {"chunk": "olleh"},
+                    "event": "on_chain_stream",
+                    "metadata": {},
+                    "name": "reverse",
+                    "tags": [],
+                },
+                {
+                    "data": {"output": "olleh"},
+                    "event": "on_chain_end",
+                    "metadata": {},
+                    "name": "reverse",
+                    "tags": [],
+                },
+            ]
+            ```
 
-        ```python title="Example: Dispatch Custom Event"
+        ```python title="Dispatch custom event"
         from langchain_core.callbacks.manager import (
             adispatch_custom_event,
         )
@@ -1428,10 +1451,13 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             input: The input to the `Runnable`.
             config: The config to use for the `Runnable`.
-            version: The version of the schema to use either `'v2'` or `'v1'`.
+            version: The version of the schema to use, either `'v2'` or `'v1'`.
+
                 Users should use `'v2'`.
+
                 `'v1'` is for backwards compatibility and will be deprecated
                 in `0.4.0`.
+
                 No default will be assigned until the API is stabilized.
                 custom events will only be surfaced in `'v2'`.
             include_names: Only include events from `Runnable` objects with matching names.
@@ -1441,6 +1467,7 @@ class Runnable(ABC, Generic[Input, Output]):
             exclude_types: Exclude events from `Runnable` objects with matching types.
             exclude_tags: Exclude events from `Runnable` objects with matching tags.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
+
                 These will be passed to `astream_log` as this implementation
                 of `astream_events` is built on top of `astream_log`.
 
@@ -1742,46 +1769,52 @@ class Runnable(ABC, Generic[Input, Output]):
             import time
             import asyncio
 
+
             def format_t(timestamp: float) -> str:
                 return datetime.fromtimestamp(timestamp, tz=timezone.utc).isoformat()
 
+
             async def test_runnable(time_to_sleep: int):
                 print(f"Runnable[{time_to_sleep}s]: starts at {format_t(time.time())}")
                 await asyncio.sleep(time_to_sleep)
                 print(f"Runnable[{time_to_sleep}s]: ends at {format_t(time.time())}")
 
+
             async def fn_start(run_obj: Runnable):
                 print(f"on start callback starts at {format_t(time.time())}")
                 await asyncio.sleep(3)
                 print(f"on start callback ends at {format_t(time.time())}")
 
+
             async def fn_end(run_obj: Runnable):
                 print(f"on end callback starts at {format_t(time.time())}")
                 await asyncio.sleep(2)
                 print(f"on end callback ends at {format_t(time.time())}")
 
+
             runnable = RunnableLambda(test_runnable).with_alisteners(
-                on_start=fn_start,
-                on_end=fn_end
+                on_start=fn_start, on_end=fn_end
             )
+
+
             async def concurrent_runs():
                 await asyncio.gather(runnable.ainvoke(2), runnable.ainvoke(3))
 
-            asyncio.run(concurrent_runs())
-            Result:
-            on start callback starts at 2025-03-01T07:05:22.875378+00:00
-            on start callback starts at 2025-03-01T07:05:22.875495+00:00
-            on start callback ends at 2025-03-01T07:05:25.878862+00:00
-            on start callback ends at 2025-03-01T07:05:25.878947+00:00
-            Runnable[2s]: starts at 2025-03-01T07:05:25.879392+00:00
-            Runnable[3s]: starts at 2025-03-01T07:05:25.879804+00:00
-            Runnable[2s]: ends at 2025-03-01T07:05:27.881998+00:00
-            on end callback starts at 2025-03-01T07:05:27.882360+00:00
-            Runnable[3s]: ends at 2025-03-01T07:05:28.881737+00:00
-            on end callback starts at 2025-03-01T07:05:28.882428+00:00
-            on end callback ends at 2025-03-01T07:05:29.883893+00:00
-            on end callback ends at 2025-03-01T07:05:30.884831+00:00
 
+            asyncio.run(concurrent_runs())
+            # Result:
+            # on start callback starts at 2025-03-01T07:05:22.875378+00:00
+            # on start callback starts at 2025-03-01T07:05:22.875495+00:00
+            # on start callback ends at 2025-03-01T07:05:25.878862+00:00
+            # on start callback ends at 2025-03-01T07:05:25.878947+00:00
+            # Runnable[2s]: starts at 2025-03-01T07:05:25.879392+00:00
+            # Runnable[3s]: starts at 2025-03-01T07:05:25.879804+00:00
+            # Runnable[2s]: ends at 2025-03-01T07:05:27.881998+00:00
+            # on end callback starts at 2025-03-01T07:05:27.882360+00:00
+            # Runnable[3s]: ends at 2025-03-01T07:05:28.881737+00:00
+            # on end callback starts at 2025-03-01T07:05:28.882428+00:00
+            # on end callback ends at 2025-03-01T07:05:29.883893+00:00
+            # on end callback ends at 2025-03-01T07:05:30.884831+00:00
             ```
         """
         return RunnableBinding(
@@ -1843,7 +1876,7 @@ class Runnable(ABC, Generic[Input, Output]):
                 `exp_base`, and `jitter` (all `float` values).
 
         Returns:
-            A new Runnable that retries the original Runnable on exceptions.
+            A new `Runnable` that retries the original `Runnable` on exceptions.
 
         Example:
             ```python
@@ -1927,7 +1960,9 @@ class Runnable(ABC, Generic[Input, Output]):
             exceptions_to_handle: A tuple of exception types to handle.
             exception_key: If `string` is specified then handled exceptions will be
                 passed to fallbacks as part of the input under the specified key.
+
                 If `None`, exceptions will not be passed to fallbacks.
+
                 If used, the base `Runnable` and its fallbacks must accept a
                 dictionary as input.
 
@@ -1963,7 +1998,9 @@ class Runnable(ABC, Generic[Input, Output]):
             exceptions_to_handle: A tuple of exception types to handle.
             exception_key: If `string` is specified then handled exceptions will be
                 passed to fallbacks as part of the input under the specified key.
+
                 If `None`, exceptions will not be passed to fallbacks.
+
                 If used, the base `Runnable` and its fallbacks must accept a
                 dictionary as input.
 
@@ -2429,10 +2466,14 @@ class Runnable(ABC, Generic[Input, Output]):
 
         `as_tool` will instantiate a `BaseTool` with a name, description, and
         `args_schema` from a `Runnable`. Where possible, schemas are inferred
-        from `runnable.get_input_schema`. Alternatively (e.g., if the
-        `Runnable` takes a dict as input and the specific dict keys are not typed),
-        the schema can be specified directly with `args_schema`. You can also
-        pass `arg_types` to just specify the required arguments and their types.
+        from `runnable.get_input_schema`.
+
+        Alternatively (e.g., if the `Runnable` takes a dict as input and the specific
+        `dict` keys are not typed), the schema can be specified directly with
+        `args_schema`.
+
+        You can also pass `arg_types` to just specify the required arguments and their
+        types.
 
         Args:
             args_schema: The schema for the tool.
@@ -2443,82 +2484,82 @@ class Runnable(ABC, Generic[Input, Output]):
         Returns:
             A `BaseTool` instance.
 
-        Typed dict input:
+        !!! example "`TypedDict` input"
 
-        ```python
-        from typing_extensions import TypedDict
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from typing_extensions import TypedDict
+            from langchain_core.runnables import RunnableLambda
 
 
-        class Args(TypedDict):
-            a: int
-            b: list[int]
+            class Args(TypedDict):
+                a: int
+                b: list[int]
 
 
-        def f(x: Args) -> str:
-            return str(x["a"] * max(x["b"]))
+            def f(x: Args) -> str:
+                return str(x["a"] * max(x["b"]))
 
 
-        runnable = RunnableLambda(f)
-        as_tool = runnable.as_tool()
-        as_tool.invoke({"a": 3, "b": [1, 2]})
-        ```
+            runnable = RunnableLambda(f)
+            as_tool = runnable.as_tool()
+            as_tool.invoke({"a": 3, "b": [1, 2]})
+            ```
 
-        `dict` input, specifying schema via `args_schema`:
+        !!! example "`dict` input, specifying schema via `args_schema`"
 
-        ```python
-        from typing import Any
-        from pydantic import BaseModel, Field
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from typing import Any
+            from pydantic import BaseModel, Field
+            from langchain_core.runnables import RunnableLambda
 
-        def f(x: dict[str, Any]) -> str:
-            return str(x["a"] * max(x["b"]))
+            def f(x: dict[str, Any]) -> str:
+                return str(x["a"] * max(x["b"]))
 
-        class FSchema(BaseModel):
-            \"\"\"Apply a function to an integer and list of integers.\"\"\"
+            class FSchema(BaseModel):
+                \"\"\"Apply a function to an integer and list of integers.\"\"\"
 
-            a: int = Field(..., description="Integer")
-            b: list[int] = Field(..., description="List of ints")
+                a: int = Field(..., description="Integer")
+                b: list[int] = Field(..., description="List of ints")
 
-        runnable = RunnableLambda(f)
-        as_tool = runnable.as_tool(FSchema)
-        as_tool.invoke({"a": 3, "b": [1, 2]})
-        ```
+            runnable = RunnableLambda(f)
+            as_tool = runnable.as_tool(FSchema)
+            as_tool.invoke({"a": 3, "b": [1, 2]})
+            ```
 
-        `dict` input, specifying schema via `arg_types`:
+        !!! example "`dict` input, specifying schema via `arg_types`"
 
-        ```python
-        from typing import Any
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from typing import Any
+            from langchain_core.runnables import RunnableLambda
 
 
-        def f(x: dict[str, Any]) -> str:
-            return str(x["a"] * max(x["b"]))
+            def f(x: dict[str, Any]) -> str:
+                return str(x["a"] * max(x["b"]))
 
 
-        runnable = RunnableLambda(f)
-        as_tool = runnable.as_tool(arg_types={"a": int, "b": list[int]})
-        as_tool.invoke({"a": 3, "b": [1, 2]})
-        ```
+            runnable = RunnableLambda(f)
+            as_tool = runnable.as_tool(arg_types={"a": int, "b": list[int]})
+            as_tool.invoke({"a": 3, "b": [1, 2]})
+            ```
 
-        String input:
+        !!! example "`str` input"
 
-        ```python
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from langchain_core.runnables import RunnableLambda
 
 
-        def f(x: str) -> str:
-            return x + "a"
+            def f(x: str) -> str:
+                return x + "a"
 
 
-        def g(x: str) -> str:
-            return x + "z"
+            def g(x: str) -> str:
+                return x + "z"
 
 
-        runnable = RunnableLambda(f) | g
-        as_tool = runnable.as_tool()
-        as_tool.invoke("b")
-        ```
+            runnable = RunnableLambda(f) | g
+            as_tool = runnable.as_tool()
+            as_tool.invoke("b")
+            ```
         """
         # Avoid circular import
         from langchain_core.tools import convert_runnable_to_tool  # noqa: PLC0415
@@ -2570,29 +2611,33 @@ class RunnableSerializable(Serializable, Runnable[Input, Output]):
         Returns:
             A new `Runnable` with the fields configured.
 
-        ```python
-        from langchain_core.runnables import ConfigurableField
-        from langchain_openai import ChatOpenAI
+        !!! example
 
-        model = ChatOpenAI(max_tokens=20).configurable_fields(
-            max_tokens=ConfigurableField(
-                id="output_token_number",
-                name="Max tokens in the output",
-                description="The maximum number of tokens in the output",
+            ```python
+            from langchain_core.runnables import ConfigurableField
+            from langchain_openai import ChatOpenAI
+
+            model = ChatOpenAI(max_tokens=20).configurable_fields(
+                max_tokens=ConfigurableField(
+                    id="output_token_number",
+                    name="Max tokens in the output",
+                    description="The maximum number of tokens in the output",
+                )
             )
-        )
 
-        # max_tokens = 20
-        print("max_tokens_20: ", model.invoke("tell me something about chess").content)
+            # max_tokens = 20
+            print(
+                "max_tokens_20: ", model.invoke("tell me something about chess").content
+            )
 
-        # max_tokens = 200
-        print(
-            "max_tokens_200: ",
-            model.with_config(configurable={"output_token_number": 200})
-            .invoke("tell me something about chess")
-            .content,
-        )
-        ```
+            # max_tokens = 200
+            print(
+                "max_tokens_200: ",
+                model.with_config(configurable={"output_token_number": 200})
+                .invoke("tell me something about chess")
+                .content,
+            )
+            ```
         """
         # Import locally to prevent circular import
         from langchain_core.runnables.configurable import (  # noqa: PLC0415
@@ -2631,29 +2676,31 @@ class RunnableSerializable(Serializable, Runnable[Input, Output]):
         Returns:
             A new `Runnable` with the alternatives configured.
 
-        ```python
-        from langchain_anthropic import ChatAnthropic
-        from langchain_core.runnables.utils import ConfigurableField
-        from langchain_openai import ChatOpenAI
+        !!! example
 
-        model = ChatAnthropic(
-            model_name="claude-3-7-sonnet-20250219"
-        ).configurable_alternatives(
-            ConfigurableField(id="llm"),
-            default_key="anthropic",
-            openai=ChatOpenAI(),
-        )
+            ```python
+            from langchain_anthropic import ChatAnthropic
+            from langchain_core.runnables.utils import ConfigurableField
+            from langchain_openai import ChatOpenAI
 
-        # uses the default model ChatAnthropic
-        print(model.invoke("which organization created you?").content)
+            model = ChatAnthropic(
+                model_name="claude-sonnet-4-5-20250929"
+            ).configurable_alternatives(
+                ConfigurableField(id="llm"),
+                default_key="anthropic",
+                openai=ChatOpenAI(),
+            )
 
-        # uses ChatOpenAI
-        print(
-            model.with_config(configurable={"llm": "openai"})
-            .invoke("which organization created you?")
-            .content
-        )
-        ```
+            # uses the default model ChatAnthropic
+            print(model.invoke("which organization created you?").content)
+
+            # uses ChatOpenAI
+            print(
+                model.with_config(configurable={"llm": "openai"})
+                .invoke("which organization created you?")
+                .content
+            )
+            ```
         """
         # Import locally to prevent circular import
         from langchain_core.runnables.configurable import (  # noqa: PLC0415
@@ -2750,6 +2797,9 @@ def _seq_output_schema(
     return last.get_output_schema(config)
 
 
+_RUNNABLE_SEQUENCE_MIN_STEPS = 2
+
+
 class RunnableSequence(RunnableSerializable[Input, Output]):
     """Sequence of `Runnable` objects, where the output of one is the input of the next.
 
@@ -2859,7 +2909,7 @@ class RunnableSequence(RunnableSerializable[Input, Output]):
             name: The name of the `Runnable`.
             first: The first `Runnable` in the sequence.
             middle: The middle `Runnable` objects in the sequence.
-            last: The last Runnable in the sequence.
+            last: The last `Runnable` in the sequence.
 
         Raises:
             ValueError: If the sequence has less than 2 steps.
@@ -2872,8 +2922,11 @@ class RunnableSequence(RunnableSerializable[Input, Output]):
                 steps_flat.extend(step.steps)
             else:
                 steps_flat.append(coerce_to_runnable(step))
-        if len(steps_flat) < 2:
-            msg = f"RunnableSequence must have at least 2 steps, got {len(steps_flat)}"
+        if len(steps_flat) < _RUNNABLE_SEQUENCE_MIN_STEPS:
+            msg = (
+                f"RunnableSequence must have at least {_RUNNABLE_SEQUENCE_MIN_STEPS} "
+                f"steps, got {len(steps_flat)}"
+            )
             raise ValueError(msg)
         super().__init__(
             first=steps_flat[0],
@@ -2904,7 +2957,7 @@ class RunnableSequence(RunnableSerializable[Input, Output]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     model_config = ConfigDict(
@@ -3500,7 +3553,7 @@ class RunnableParallel(RunnableSerializable[Input, dict[str, Any]]):
 
     Returns a mapping of their outputs.
 
-    `RunnableParallel` is one of the two main composition primitives for the LCEL,
+    `RunnableParallel` is one of the two main composition primitives,
     alongside `RunnableSequence`. It invokes `Runnable`s concurrently, providing the
     same input to each.
 
@@ -3610,7 +3663,7 @@ class RunnableParallel(RunnableSerializable[Input, dict[str, Any]]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -3668,6 +3721,12 @@ class RunnableParallel(RunnableSerializable[Input, dict[str, Any]]):
             == "object"
             for s in self.steps__.values()
         ):
+            for step in self.steps__.values():
+                fields = step.get_input_schema(config).model_fields
+                root_field = fields.get("root")
+                if root_field is not None and root_field.annotation != Any:
+                    return super().get_input_schema(config)
+
             # This is correct, but pydantic typings/mypy don't think so.
             return create_model_v2(
                 self.get_name("Input"),
@@ -4477,7 +4536,7 @@ class RunnableLambda(Runnable[Input, Output]):
             # on itemgetter objects, so we have to parse the repr
             items = str(func).replace("operator.itemgetter(", "")[:-1].split(", ")
             if all(
-                item[0] == "'" and item[-1] == "'" and len(item) > 2 for item in items
+                item[0] == "'" and item[-1] == "'" and item != "''" for item in items
             ):
                 fields = {item[1:-1]: (Any, ...) for item in items}
                 # It's a dict, lol
@@ -5139,7 +5198,7 @@ class RunnableEachBase(RunnableSerializable[list[Input], list[Output]]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -5322,7 +5381,7 @@ class RunnableEach(RunnableEachBase[Input, Output]):
 
 
 class RunnableBindingBase(RunnableSerializable[Input, Output]):  # type: ignore[no-redef]
-    """`Runnable` that delegates calls to another `Runnable` with a set of kwargs.
+    """`Runnable` that delegates calls to another `Runnable` with a set of `**kwargs`.
 
     Use only if creating a new `RunnableBinding` subclass with different `__init__`
     args.
@@ -5462,7 +5521,7 @@ class RunnableBindingBase(RunnableSerializable[Input, Output]):  # type: ignore[
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -5752,7 +5811,7 @@ class RunnableBinding(RunnableBindingBase[Input, Output]):  # type: ignore[no-re
         ```python
         # Create a Runnable binding that invokes the chat model with the
         # additional kwarg `stop=['-']` when running it.
-        from langchain_community.chat_models import ChatOpenAI
+        from langchain_openai import ChatOpenAI
 
         model = ChatOpenAI()
         model.invoke('Say "Parrot-MAGIC"', stop=["-"])  # Should return `Parrot`

libs/core/langchain_core/runnables/branch.py

@@ -36,11 +36,13 @@ from langchain_core.runnables.utils import (
     get_unique_config_specs,
 )
 
+_MIN_BRANCHES = 2
+
 
 class RunnableBranch(RunnableSerializable[Input, Output]):
-    """Runnable that selects which branch to run based on a condition.
+    """`Runnable` that selects which branch to run based on a condition.
 
-    The Runnable is initialized with a list of `(condition, Runnable)` pairs and
+    The `Runnable` is initialized with a list of `(condition, Runnable)` pairs and
     a default branch.
 
     When operating on an input, the first condition that evaluates to True is
@@ -86,12 +88,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
                 Defaults a `Runnable` to run if no condition is met.
 
         Raises:
-            ValueError: If the number of branches is less than 2.
+            ValueError: If the number of branches is less than `2`.
             TypeError: If the default branch is not `Runnable`, `Callable` or `Mapping`.
-            TypeError: If a branch is not a tuple or list.
-            ValueError: If a branch is not of length 2.
+            TypeError: If a branch is not a `tuple` or `list`.
+            ValueError: If a branch is not of length `2`.
         """
-        if len(branches) < 2:
+        if len(branches) < _MIN_BRANCHES:
             msg = "RunnableBranch requires at least two branches"
             raise ValueError(msg)
 
@@ -118,7 +120,7 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
                 )
                 raise TypeError(msg)
 
-            if len(branch) != 2:
+            if len(branch) != _MIN_BRANCHES:
                 msg = (
                     f"RunnableBranch branches must be "
                     f"tuples or lists of length 2, not {len(branch)}"
@@ -140,7 +142,7 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -187,12 +189,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
     def invoke(
         self, input: Input, config: RunnableConfig | None = None, **kwargs: Any
     ) -> Output:
-        """First evaluates the condition, then delegate to true or false branch.
+        """First evaluates the condition, then delegate to `True` or `False` branch.
 
         Args:
-            input: The input to the Runnable.
-            config: The configuration for the Runnable.
-            **kwargs: Additional keyword arguments to pass to the Runnable.
+            input: The input to the `Runnable`.
+            config: The configuration for the `Runnable`.
+            **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
         Returns:
             The output of the branch that was run.
@@ -297,12 +299,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
         config: RunnableConfig | None = None,
         **kwargs: Any | None,
     ) -> Iterator[Output]:
-        """First evaluates the condition, then delegate to true or false branch.
+        """First evaluates the condition, then delegate to `True` or `False` branch.
 
         Args:
-            input: The input to the Runnable.
-            config: The configuration for the Runnable.
-            **kwargs: Additional keyword arguments to pass to the Runnable.
+            input: The input to the `Runnable`.
+            config: The configuration for the `Runnable`.
+            **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
         Yields:
             The output of the branch that was run.
@@ -381,12 +383,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
         config: RunnableConfig | None = None,
         **kwargs: Any | None,
     ) -> AsyncIterator[Output]:
-        """First evaluates the condition, then delegate to true or false branch.
+        """First evaluates the condition, then delegate to `True` or `False` branch.
 
         Args:
-            input: The input to the Runnable.
-            config: The configuration for the Runnable.
-            **kwargs: Additional keyword arguments to pass to the Runnable.
+            input: The input to the `Runnable`.
+            config: The configuration for the `Runnable`.
+            **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
         Yields:
             The output of the branch that was run.

libs/core/langchain_core/runnables/config.py

@@ -47,54 +47,59 @@ class EmptyDict(TypedDict, total=False):
 
 
 class RunnableConfig(TypedDict, total=False):
-    """Configuration for a Runnable."""
+    """Configuration for a `Runnable`.
+
+    See the [reference docs](https://reference.langchain.com/python/langchain_core/runnables/#langchain_core.runnables.RunnableConfig)
+    for more details.
+    """
 
     tags: list[str]
-    """
-    Tags for this call and any sub-calls (eg. a Chain calling an LLM).
+    """Tags for this call and any sub-calls (e.g. a Chain calling an LLM).
+
     You can use these to filter calls.
     """
 
     metadata: dict[str, Any]
-    """
-    Metadata for this call and any sub-calls (eg. a Chain calling an LLM).
+    """Metadata for this call and any sub-calls (e.g. a Chain calling an LLM).
+
     Keys should be strings, values should be JSON-serializable.
     """
 
     callbacks: Callbacks
-    """
-    Callbacks for this call and any sub-calls (eg. a Chain calling an LLM).
+    """Callbacks for this call and any sub-calls (e.g. a Chain calling an LLM).
+
     Tags are passed to all callbacks, metadata is passed to handle*Start callbacks.
     """
 
     run_name: str
-    """
-    Name for the tracer run for this call. Defaults to the name of the class.
-    """
+    """Name for the tracer run for this call.
+
+    Defaults to the name of the class."""
 
     max_concurrency: int | None
-    """
-    Maximum number of parallel calls to make. If not provided, defaults to
-    `ThreadPoolExecutor`'s default.
+    """Maximum number of parallel calls to make.
+
+    If not provided, defaults to `ThreadPoolExecutor`'s default.
     """
 
     recursion_limit: int
-    """
-    Maximum number of times a call can recurse. If not provided, defaults to `25`.
+    """Maximum number of times a call can recurse.
+
+    If not provided, defaults to `25`.
     """
 
     configurable: dict[str, Any]
-    """
-    Runtime values for attributes previously made configurable on this `Runnable`,
+    """Runtime values for attributes previously made configurable on this `Runnable`,
     or sub-Runnables, through `configurable_fields` or `configurable_alternatives`.
+
     Check `output_schema` for a description of the attributes that have been made
     configurable.
     """
 
     run_id: uuid.UUID | None
-    """
-    Unique identifier for the tracer run for this call. If not provided, a new UUID
-    will be generated.
+    """Unique identifier for the tracer run for this call.
+
+    If not provided, a new UUID will be generated.
     """
 
 

libs/core/langchain_core/runnables/configurable.py

@@ -1,4 +1,4 @@
-"""Runnables that can be dynamically configured."""
+"""`Runnable` objects that can be dynamically configured."""
 
 from __future__ import annotations
 
@@ -47,14 +47,14 @@ if TYPE_CHECKING:
 
 
 class DynamicRunnable(RunnableSerializable[Input, Output]):
-    """Serializable Runnable that can be dynamically configured.
+    """Serializable `Runnable` that can be dynamically configured.
 
-    A DynamicRunnable should be initiated using the `configurable_fields` or
-    `configurable_alternatives` method of a Runnable.
+    A `DynamicRunnable` should be initiated using the `configurable_fields` or
+    `configurable_alternatives` method of a `Runnable`.
     """
 
     default: RunnableSerializable[Input, Output]
-    """The default Runnable to use."""
+    """The default `Runnable` to use."""
 
     config: RunnableConfig | None = None
     """The configuration to use."""
@@ -66,7 +66,7 @@ class DynamicRunnable(RunnableSerializable[Input, Output]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -120,13 +120,13 @@ class DynamicRunnable(RunnableSerializable[Input, Output]):
     def prepare(
         self, config: RunnableConfig | None = None
     ) -> tuple[Runnable[Input, Output], RunnableConfig]:
-        """Prepare the Runnable for invocation.
+        """Prepare the `Runnable` for invocation.
 
         Args:
             config: The configuration to use.
 
         Returns:
-            The prepared Runnable and configuration.
+            The prepared `Runnable` and configuration.
         """
         runnable: Runnable[Input, Output] = self
         while isinstance(runnable, DynamicRunnable):
@@ -316,12 +316,12 @@ class DynamicRunnable(RunnableSerializable[Input, Output]):
 
 
 class RunnableConfigurableFields(DynamicRunnable[Input, Output]):
-    """Runnable that can be dynamically configured.
+    """`Runnable` that can be dynamically configured.
 
-    A RunnableConfigurableFields should be initiated using the
-    `configurable_fields` method of a Runnable.
+    A `RunnableConfigurableFields` should be initiated using the
+    `configurable_fields` method of a `Runnable`.
 
-    Here is an example of using a RunnableConfigurableFields with LLMs:
+    Here is an example of using a `RunnableConfigurableFields` with LLMs:
 
         ```python
         from langchain_core.prompts import PromptTemplate
@@ -348,7 +348,7 @@ class RunnableConfigurableFields(DynamicRunnable[Input, Output]):
         chain.invoke({"x": 0}, config={"configurable": {"temperature": 0.9}})
         ```
 
-    Here is an example of using a RunnableConfigurableFields with HubRunnables:
+    Here is an example of using a `RunnableConfigurableFields` with `HubRunnables`:
 
         ```python
         from langchain_core.prompts import PromptTemplate
@@ -380,7 +380,7 @@ class RunnableConfigurableFields(DynamicRunnable[Input, Output]):
 
     @property
     def config_specs(self) -> list[ConfigurableFieldSpec]:
-        """Get the configuration specs for the RunnableConfigurableFields.
+        """Get the configuration specs for the `RunnableConfigurableFields`.
 
         Returns:
             The configuration specs.
@@ -473,13 +473,13 @@ _enums_for_spec_lock = threading.Lock()
 
 
 class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
-    """Runnable that can be dynamically configured.
+    """`Runnable` that can be dynamically configured.
 
-    A RunnableConfigurableAlternatives should be initiated using the
-    `configurable_alternatives` method of a Runnable or can be
+    A `RunnableConfigurableAlternatives` should be initiated using the
+    `configurable_alternatives` method of a `Runnable` or can be
     initiated directly as well.
 
-    Here is an example of using a RunnableConfigurableAlternatives that uses
+    Here is an example of using a `RunnableConfigurableAlternatives` that uses
     alternative prompts to illustrate its functionality:
 
         ```python
@@ -506,7 +506,7 @@ class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
         chain.with_config(configurable={"prompt": "poem"}).invoke({"topic": "bears"})
         ```
 
-    Equivalently, you can initialize RunnableConfigurableAlternatives directly
+    Equivalently, you can initialize `RunnableConfigurableAlternatives` directly
     and use in LCEL in the same way:
 
         ```python
@@ -531,7 +531,7 @@ class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
     """
 
     which: ConfigurableField
-    """The ConfigurableField to use to choose between alternatives."""
+    """The `ConfigurableField` to use to choose between alternatives."""
 
     alternatives: dict[
         str,
@@ -544,8 +544,9 @@ class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
 
     prefix_keys: bool
     """Whether to prefix configurable fields of each alternative with a namespace
-    of the form <which.id>==<alternative_key>, eg. a key named "temperature" used by
-    the alternative named "gpt3" becomes "model==gpt3/temperature"."""
+    of the form <which.id>==<alternative_key>, e.g. a key named "temperature" used by
+    the alternative named "gpt3" becomes "model==gpt3/temperature".
+    """
 
     @property
     @override
@@ -638,24 +639,24 @@ class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
 
 
 def _strremoveprefix(s: str, prefix: str) -> str:
-    """str.removeprefix() is only available in Python 3.9+."""
+    """`str.removeprefix()` is only available in Python 3.9+."""
     return s.replace(prefix, "", 1) if s.startswith(prefix) else s
 
 
 def prefix_config_spec(
     spec: ConfigurableFieldSpec, prefix: str
 ) -> ConfigurableFieldSpec:
-    """Prefix the id of a ConfigurableFieldSpec.
+    """Prefix the id of a `ConfigurableFieldSpec`.
 
-    This is useful when a RunnableConfigurableAlternatives is used as a
-    ConfigurableField of another RunnableConfigurableAlternatives.
+    This is useful when a `RunnableConfigurableAlternatives` is used as a
+    `ConfigurableField` of another `RunnableConfigurableAlternatives`.
 
     Args:
-        spec: The ConfigurableFieldSpec to prefix.
+        spec: The `ConfigurableFieldSpec` to prefix.
         prefix: The prefix to add.
 
     Returns:
-        The prefixed ConfigurableFieldSpec.
+        The prefixed `ConfigurableFieldSpec`.
     """
     return (
         ConfigurableFieldSpec(
@@ -677,15 +678,15 @@ def make_options_spec(
 ) -> ConfigurableFieldSpec:
     """Make options spec.
 
-    Make a ConfigurableFieldSpec for a ConfigurableFieldSingleOption or
-    ConfigurableFieldMultiOption.
+    Make a `ConfigurableFieldSpec` for a `ConfigurableFieldSingleOption` or
+    `ConfigurableFieldMultiOption`.
 
     Args:
-        spec: The ConfigurableFieldSingleOption or ConfigurableFieldMultiOption.
+        spec: The `ConfigurableFieldSingleOption` or `ConfigurableFieldMultiOption`.
         description: The description to use if the spec does not have one.
 
     Returns:
-        The ConfigurableFieldSpec.
+        The `ConfigurableFieldSpec`.
     """
     with _enums_for_spec_lock:
         if enum := _enums_for_spec.get(spec):

libs/core/langchain_core/runnables/fallbacks.py

@@ -35,20 +35,20 @@ if TYPE_CHECKING:
 
 
 class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
-    """Runnable that can fallback to other Runnables if it fails.
+    """`Runnable` that can fallback to other `Runnable`s if it fails.
 
     External APIs (e.g., APIs for a language model) may at times experience
     degraded performance or even downtime.
 
-    In these cases, it can be useful to have a fallback Runnable that can be
-    used in place of the original Runnable (e.g., fallback to another LLM provider).
+    In these cases, it can be useful to have a fallback `Runnable` that can be
+    used in place of the original `Runnable` (e.g., fallback to another LLM provider).
 
-    Fallbacks can be defined at the level of a single Runnable, or at the level
-    of a chain of Runnables. Fallbacks are tried in order until one succeeds or
+    Fallbacks can be defined at the level of a single `Runnable`, or at the level
+    of a chain of `Runnable`s. Fallbacks are tried in order until one succeeds or
     all fail.
 
     While you can instantiate a `RunnableWithFallbacks` directly, it is usually
-    more convenient to use the `with_fallbacks` method on a Runnable.
+    more convenient to use the `with_fallbacks` method on a `Runnable`.
 
     Example:
         ```python
@@ -87,7 +87,7 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
     """
 
     runnable: Runnable[Input, Output]
-    """The Runnable to run first."""
+    """The `Runnable` to run first."""
     fallbacks: Sequence[Runnable[Input, Output]]
     """A sequence of fallbacks to try."""
     exceptions_to_handle: tuple[type[BaseException], ...] = (Exception,)
@@ -97,9 +97,12 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
     """
     exception_key: str | None = None
     """If `string` is specified then handled exceptions will be passed to fallbacks as
-        part of the input under the specified key. If `None`, exceptions
-        will not be passed to fallbacks. If used, the base Runnable and its fallbacks
-        must accept a dictionary as input."""
+    part of the input under the specified key.
+
+    If `None`, exceptions will not be passed to fallbacks.
+
+    If used, the base `Runnable` and its fallbacks must accept a dictionary as input.
+    """
 
     model_config = ConfigDict(
         arbitrary_types_allowed=True,
@@ -137,7 +140,7 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -152,10 +155,10 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
 
     @property
     def runnables(self) -> Iterator[Runnable[Input, Output]]:
-        """Iterator over the Runnable and its fallbacks.
+        """Iterator over the `Runnable` and its fallbacks.
 
         Yields:
-            The Runnable then its fallbacks.
+            The `Runnable` then its fallbacks.
         """
         yield self.runnable
         yield from self.fallbacks
@@ -589,14 +592,14 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
         await run_manager.on_chain_end(output)
 
     def __getattr__(self, name: str) -> Any:
-        """Get an attribute from the wrapped Runnable and its fallbacks.
+        """Get an attribute from the wrapped `Runnable` and its fallbacks.
 
         Returns:
-            If the attribute is anything other than a method that outputs a Runnable,
-            returns getattr(self.runnable, name). If the attribute is a method that
-            does return a new Runnable (e.g. model.bind_tools([...]) outputs a new
-            RunnableBinding) then self.runnable and each of the runnables in
-            self.fallbacks is replaced with getattr(x, name).
+            If the attribute is anything other than a method that outputs a `Runnable`,
+            returns `getattr(self.runnable, name)`. If the attribute is a method that
+            does return a new `Runnable` (e.g. `model.bind_tools([...])` outputs a new
+            `RunnableBinding`) then `self.runnable` and each of the runnables in
+            `self.fallbacks` is replaced with `getattr(x, name)`.
 
         Example:
             ```python
@@ -604,7 +607,7 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
             from langchain_anthropic import ChatAnthropic
 
             gpt_4o = ChatOpenAI(model="gpt-4o")
-            claude_3_sonnet = ChatAnthropic(model="claude-3-7-sonnet-20250219")
+            claude_3_sonnet = ChatAnthropic(model="claude-sonnet-4-5-20250929")
             model = gpt_4o.with_fallbacks([claude_3_sonnet])
 
             model.model_name
@@ -618,7 +621,6 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
                 runnable=RunnableBinding(bound=ChatOpenAI(...), kwargs={"tools": [...]}),
                 fallbacks=[RunnableBinding(bound=ChatAnthropic(...), kwargs={"tools": [...]})],
             )
-
             ```
         """  # noqa: E501
         attr = getattr(self.runnable, name)

libs/core/langchain_core/runnables/graph.py

@@ -4,7 +4,6 @@ from __future__ import annotations
 
 import inspect
 from collections import defaultdict
-from collections.abc import Callable
 from dataclasses import dataclass, field
 from enum import Enum
 from typing import (
@@ -22,7 +21,7 @@ from langchain_core.runnables.base import Runnable, RunnableSerializable
 from langchain_core.utils.pydantic import _IgnoreUnserializable, is_basemodel_subclass
 
 if TYPE_CHECKING:
-    from collections.abc import Sequence
+    from collections.abc import Callable, Sequence
 
     from pydantic import BaseModel
 
@@ -132,7 +131,7 @@ class Branch(NamedTuple):
     condition: Callable[..., str]
     """A callable that returns a string representation of the condition."""
     ends: dict[str, str] | None
-    """Optional dictionary of end node ids for the branches. """
+    """Optional dictionary of end node IDs for the branches. """
 
 
 class CurveStyle(Enum):
@@ -642,6 +641,7 @@ class Graph:
         retry_delay: float = 1.0,
         frontmatter_config: dict[str, Any] | None = None,
         base_url: str | None = None,
+        proxies: dict[str, str] | None = None,
     ) -> bytes:
         """Draw the graph as a PNG image using Mermaid.
 
@@ -674,11 +674,10 @@ class Graph:
                 }
                 ```
             base_url: The base URL of the Mermaid server for rendering via API.
-
+            proxies: HTTP/HTTPS proxies for requests (e.g. `{"http": "http://127.0.0.1:7890"}`).
 
         Returns:
             The PNG image as bytes.
-
         """
         # Import locally to prevent circular import
         from langchain_core.runnables.graph_mermaid import (  # noqa: PLC0415
@@ -699,6 +698,7 @@ class Graph:
             padding=padding,
             max_retries=max_retries,
             retry_delay=retry_delay,
+            proxies=proxies,
             base_url=base_url,
         )
 
@@ -706,8 +706,10 @@ class Graph:
 def _first_node(graph: Graph, exclude: Sequence[str] = ()) -> Node | None:
     """Find the single node that is not a target of any edge.
 
-    Exclude nodes/sources with ids in the exclude list.
+    Exclude nodes/sources with IDs in the exclude list.
+
     If there is no such node, or there are multiple, return `None`.
+
     When drawing the graph, this node would be the origin.
     """
     targets = {edge.target for edge in graph.edges if edge.source not in exclude}
@@ -722,8 +724,10 @@ def _first_node(graph: Graph, exclude: Sequence[str] = ()) -> Node | None:
 def _last_node(graph: Graph, exclude: Sequence[str] = ()) -> Node | None:
     """Find the single node that is not a source of any edge.
 
-    Exclude nodes/targets with ids in the exclude list.
+    Exclude nodes/targets with IDs in the exclude list.
+
     If there is no such node, or there are multiple, return `None`.
+
     When drawing the graph, this node would be the destination.
     """
     sources = {edge.source for edge in graph.edges if edge.target not in exclude}

libs/core/langchain_core/runnables/graph_ascii.py

@@ -7,7 +7,6 @@ from __future__ import annotations
 
 import math
 import os
-from collections.abc import Mapping, Sequence
 from typing import TYPE_CHECKING, Any
 
 try:
@@ -20,6 +19,8 @@ except ImportError:
     _HAS_GRANDALF = False
 
 if TYPE_CHECKING:
+    from collections.abc import Mapping, Sequence
+
     from langchain_core.runnables.graph import Edge as LangEdge