Commit Graph

1262 Commits

Author SHA1 Message Date
ccurme
94139ffcd3
openai[patch]: format system content blocks for Responses API (#31096)
```python
from langchain_core.messages import HumanMessage, SystemMessage
from langchain_openai import ChatOpenAI


llm = ChatOpenAI(model="gpt-4.1", use_responses_api=True)

messages = [
    SystemMessage("test"),                                   # Works
    HumanMessage("test"),                                    # Works
    SystemMessage([{"type": "text", "text": "test"}]),       # Bug in this case
    HumanMessage([{"type": "text", "text": "test"}]),        # Works
    SystemMessage([{"type": "input_text", "text": "test"}])  # Works
]

llm._get_request_payload(messages)
```
2025-05-02 15:22:30 +00:00
ccurme
26ad239669
core, openai[patch]: prefer provider-assigned IDs when aggregating message chunks (#31080)
When aggregating AIMessageChunks in a stream, core prefers the leftmost
non-null ID. This is problematic because:
- Core assigns IDs when they are null to `f"run-{run_manager.run_id}"`
- The desired meaningful ID might not be available until midway through
the stream, as is the case for the OpenAI Responses API.

For the OpenAI Responses API, we assign message IDs to the top-level
`AIMessage.id`. This works in `.(a)invoke`, but during `.(a)stream` the
IDs get overwritten by the defaults assigned in langchain-core. These
IDs
[must](https://community.openai.com/t/how-to-solve-badrequesterror-400-item-rs-of-type-reasoning-was-provided-without-its-required-following-item-error-in-responses-api/1151686/9)
be available on the AIMessage object to support passing reasoning items
back to the API (e.g., if not using OpenAI's `previous_response_id`
feature). We could add them elsewhere, but seeing as we've already made
the decision to store them in `.id` during `.(a)invoke`, addressing the
issue in core lets us fix the problem with no interface changes.
2025-05-02 11:18:18 -04:00
ccurme
c51eadd54f
openai[patch]: propagate service_tier to response metadata (#31089) 2025-05-01 13:50:48 -04:00
ccurme
6110c3ffc5
openai[patch]: release 0.3.15 (#31087) 2025-05-01 09:22:30 -04:00
Ben Gladwell
da59eb7eb4
anthropic: Allow kwargs to pass through when counting tokens (#31082)
- **Description:** `ChatAnthropic.get_num_tokens_from_messages` does not
currently receive `kwargs` and pass those on to
`self._client.beta.messages.count_tokens`. This is a problem if you need
to pass specific options to `count_tokens`, such as the `thinking`
option. This PR fixes that.
- **Issue:** N/A
- **Dependencies:** None
- **Twitter handle:** @bengladwell

Co-authored-by: ccurme <chester.curme@gmail.com>
2025-04-30 17:56:22 -04:00
Really Him
918c950737
DOCS: partners/chroma: Fix documentation around chroma query filter syntax (#31058)
Thank you for contributing to LangChain!

- [x] **PR title**: "package: description"
- Where "package" is whichever of langchain, community, core, etc. is
being modified. Use "docs: ..." for purely docs changes, "infra: ..."
for CI changes.
  - Example: "community: add foobar LLM"

**Description**:
* Starting to put together some PR's to fix the typing around
`langchain-chroma` `filter` and `where_document` query filtering, as
mentioned:

https://github.com/langchain-ai/langchain/issues/30879
https://github.com/langchain-ai/langchain/issues/30507

The typing of `dict[str, str]` is on the one hand too restrictive (marks
valid filter expressions as ill-typed) and also too permissive (allows
illegal filter expressions). That's not what this PR addresses though.
This PR just removes from the documentation some examples of filters
that are illegal, and also syntactically incorrect: (a) dictionaries
with keys like `$contains` but the key is missing quotation marks; (b)
dictionaries with multiple entries - this is illegal in Chroma filter
syntax and will raise an exception. (`{"foo": "bar", "qux": "baz"}`).
Filter dictionaries in Chroma must have one and one key only. Again this
is just the documentation issue, which is the lowest hanging fruit. I
also think we need to update the types for `filter` and `where_document`
to be (at the very least `dict[str, Any]`), or, since we have access to
Chroma's types, they should be `Where` and `WhereDocument` types. This
has a wider blast radius though, so I'm starting small.

This PR does not fix the issues mentioned above, it's just starting to
get the ball rolling, and cleaning up the documentation.



- [x] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/

Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.

If no one reviews your PR within a few days, please @-mention one of
baskaryan, eyurtsev, ccurme, vbarda, hwchase17.

---------

Co-authored-by: Really Him <hesereallyhim@proton.me>
2025-04-30 17:51:07 -04:00
ccurme
bdb7c4a8b3
huggingface: fix embeddings return type (#31072)
Integration tests failing

cc @hanouticelina
2025-04-29 18:45:04 +00:00
célina
868f07f8f4
partners: (langchain-huggingface) Chat Models - Integrate Hugging Face Inference Providers and remove deprecated code (#30733)
Hi there, I'm Célina from 🤗,
This PR introduces support for Hugging Face's serverless Inference
Providers (documentation
[here](https://huggingface.co/docs/inference-providers/index)), allowing
users to specify different providers for chat completion and text
generation tasks.

This PR also removes the usage of `InferenceClient.post()` method in
`HuggingFaceEndpoint`, in favor of the task-specific `text_generation`
method. `InferenceClient.post()` is deprecated and will be removed in
`huggingface_hub v0.31.0`.

---
## Changes made
- bumped the minimum required version of the `huggingface-hub` package
to ensure compatibility with the latest API usage.
- added a `provider` field to `HuggingFaceEndpoint`, enabling users to
select the inference provider (e.g., 'cerebras', 'together',
'fireworks-ai'). Defaults to `hf-inference` (HF Inference API).
- replaced the deprecated `InferenceClient.post()` call in
`HuggingFaceEndpoint` with the task-specific `text_generation` method
for future-proofing, `post()` will be removed in huggingface-hub
v0.31.0.
- updated the `ChatHuggingFace` component:
    - added async and streaming support.
    - added support for tool calling.
- exposed underlying chat completion parameters for more granular
control.
- Added integration tests for `ChatHuggingFace` and updated the
corresponding unit tests.

  All changes are backward compatible.

---------

Co-authored-by: ccurme <chester.curme@gmail.com>
2025-04-29 09:53:14 -04:00
Sydney Runkle
7e926520d5
packaging: remove Python upper bound for langchain and co libs (#31025)
Follow up to https://github.com/langchain-ai/langsmith-sdk/pull/1696,
I've bumped the `langsmith` version where applicable in `uv.lock`.

Type checking problems here because deps have been updated in
`pyproject.toml` and `uv lock` hasn't been run - we should enforce that
in the future - goes with the other dependabot todos :).
2025-04-28 14:44:28 -04:00
Sydney Runkle
d614842d23
ci: temporarily run chroma on 3.12 for CI (#31056)
Waiting on a fix for https://github.com/chroma-core/chroma/issues/4382
2025-04-28 13:20:37 -04:00
湛露先生
5fb8fd863a
langchain_openai: clean duplicate code for openai embedding. (#30872)
The `_chunk_size` has not changed by method `self._tokenize`, So i think
these is duplicate code.

Signed-off-by: zhanluxianshen <zhanluxianshen@163.com>
2025-04-27 15:07:41 -04:00
ccurme
a60fd06784
docs: document OpenAI flex processing (#31023)
Following https://github.com/langchain-ai/langchain/pull/31005
2025-04-25 15:10:25 -04:00
ccurme
629b7a5a43
openai[patch]: add explicit attribute for service tier (#31005) 2025-04-25 18:38:23 +00:00
ccurme
a7903280dd
openai[patch]: delete redundant tests (#31004)
These are covered by standard tests.
2025-04-24 17:56:32 +00:00
ccurme
10a9c24dae
openai: fix streaming reasoning without summaries (#30999)
Following https://github.com/langchain-ai/langchain/pull/30909: need to
retain "empty" reasoning output when streaming, e.g.,
```python
{'id': 'rs_...', 'summary': [], 'type': 'reasoning'}
```
Tested by existing integration tests, which are currently failing.
2025-04-24 16:01:45 +00:00
ccurme
faef3e5d50
core, standard-tests: support PDF and audio input in Chat Completions format (#30979)
Chat models currently implement support for:
- images in OpenAI Chat Completions format
- other multimodal types (e.g., PDF and audio) in a cross-provider
[standard
format](https://python.langchain.com/docs/how_to/multimodal_inputs/)

Here we update core to extend support to PDF and audio input in Chat
Completions format. **If an OAI-format PDF or audio content block is
passed into any chat model, it will be transformed to the LangChain
standard format**. We assume that any chat model supporting OAI-format
PDF or audio has implemented support for the standard format.
2025-04-23 18:32:51 +00:00
ccurme
4bc70766b5
core, openai: support standard multi-modal blocks in convert_to_openai_messages (#30968) 2025-04-23 11:20:44 -04:00
ccurme
e4877e5ef1
fireworks: release 0.3.0 (#30977) 2025-04-23 10:08:38 -04:00
ccurme
eedda164c6
fireworks[minor]: remove default model and temperature (#30965)
`mixtral-8x-7b-instruct` was recently retired from Fireworks Serverless.

Here we remove the default model altogether, so that the model must be
explicitly specified on init:
```python
ChatFireworks(model="accounts/fireworks/models/llama-v3p1-70b-instruct")  # for example
```

We also set a null default for `temperature`, which previously defaulted
to 0.0. This parameter will no longer be included in request payloads
unless it is explicitly provided.
2025-04-22 15:58:58 -04:00
ccurme
a7c1bccd6a
openai[patch]: remove xfails from image token counting tests (#30963)
These appear to be passing again.
2025-04-22 15:55:33 +00:00
Dmitrii Rashchenko
a43df006de
Support of openai reasoning summary streaming (#30909)
**langchain_openai: Support of reasoning summary streaming**

**Description:**
OpenAI API now supports streaming reasoning summaries for reasoning
models (o1, o3, o3-mini, o4-mini). More info about it:
https://platform.openai.com/docs/guides/reasoning#reasoning-summaries

It is supported only in Responses API (not Completion API), so you need
to create LangChain Open AI model as follows to support reasoning
summaries streaming:

```
llm = ChatOpenAI(
    model="o4-mini", # also o1, o3, o3-mini support reasoning streaming
    use_responses_api=True,  # reasoning streaming works only with responses api, not completion api
    model_kwargs={
        "reasoning": {
            "effort": "high",  # also "low" and "medium" supported
            "summary": "auto"  # some models support "concise" summary, some "detailed", but auto will always work
        }
    }
)
```

Now, if you stream events from llm:

```
async for event in llm.astream_events(prompt, version="v2"):
    print(event)
```

or

```
for chunk in llm.stream(prompt):
    print (chunk)
```

OpenAI API will send you new types of events:
`response.reasoning_summary_text.added`
`response.reasoning_summary_text.delta`
`response.reasoning_summary_text.done`

These events are new, so they were ignored. So I have added support of
these events in function `_convert_responses_chunk_to_generation_chunk`,
so reasoning chunks or full reasoning added to the chunk
additional_kwargs.

Example of how this reasoning summary may be printed:

```
    async for event in llm.astream_events(prompt, version="v2"):
        if event["event"] == "on_chat_model_stream":
            chunk: AIMessageChunk = event["data"]["chunk"]
            if "reasoning_summary_chunk" in chunk.additional_kwargs:
                print(chunk.additional_kwargs["reasoning_summary_chunk"], end="")
            elif "reasoning_summary" in chunk.additional_kwargs:
                print("\n\nFull reasoning step summary:", chunk.additional_kwargs["reasoning_summary"])
            elif chunk.content and chunk.content[0]["type"] == "text":
                print(chunk.content[0]["text"], end="")
```

or

```
    for chunk in llm.stream(prompt):
        if "reasoning_summary_chunk" in chunk.additional_kwargs:
            print(chunk.additional_kwargs["reasoning_summary_chunk"], end="")
        elif "reasoning_summary" in chunk.additional_kwargs:
            print("\n\nFull reasoning step summary:", chunk.additional_kwargs["reasoning_summary"])
        elif chunk.content and chunk.content[0]["type"] == "text":
            print(chunk.content[0]["text"], end="")
```

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-04-22 14:51:13 +00:00
ccurme
920d504e47
fireworks[patch]: update model in LLM integration tests (#30951)
`mixtral-8x7b-instruct` has been retired.
2025-04-21 17:53:27 +00:00
Ahmed Tammaa
589bc19890
anthropic[patch]: make description optional on AnthropicTool (#30935)
PR Summary

This change adds a fallback in ChatAnthropic.with_structured_output() to
handle Pydantic models that don’t include a docstring. Without it,
calling:
```py
from pydantic import BaseModel
from langchain_anthropic import ChatAnthropic

class SampleModel(BaseModel):
    sample_field: str

llm = ChatAnthropic(
    model="claude-3-7-sonnet-latest"
).with_structured_output(SampleModel.model_json_schema())

llm.invoke("test")
```
will raise a
```
KeyError: 'description'
```
because Pydantic omits the description field when no docstring is
present.

This issue doesn’t occur when using ChatOpenAI or if you add a docstring
to the model:
```py
from pydantic import BaseModel
from langchain_openai import ChatOpenAI

class SampleModel(BaseModel):
    """Schema for sample_field output."""
    sample_field: str

llm = ChatOpenAI(
    model="gpt-4o-mini"
).with_structured_output(SampleModel.model_json_schema())

llm.invoke("test")
```

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-04-21 10:44:39 -04:00
Aubrey Ford
b344f34635
partners/openai: OpenAIEmbeddings not respecting chunk_size argument (#30757)
When calling `embed_documents` and providing a `chunk_size` argument,
that argument is ignored when `OpenAIEmbeddings` is instantiated with
its default configuration (where `check_embedding_ctx_length=True`).

`_get_len_safe_embeddings` specifies a `chunk_size` parameter but it's
not being passed through in `embed_documents`, which is its only caller.
This appears to be an oversight, especially given that the
`_get_len_safe_embeddings` docstring states it should respect "the set
embedding context length and chunk size."

Developers typically expect method parameters to take effect (also, take
precedence) when explicitly provided, especially when instantiating
using defaults. I was confused as to why my API calls were being
rejected regardless of the chunk size I provided.

This bug also exists in langchain_community package. I can add that to
this PR if requested otherwise I will create a new one once this passes.
2025-04-18 15:27:27 -04:00
Konsti-s
017c8079e1
partners: ChatAnthropic supports urls (#30809)
**Description:**
partners-anthropic: ChatAnthropic supports b64 and urls in the
part[image_url][url] message variable

**Issue**:
ChatAnthropic right now only supports b64 encoded images in the
part[image_url][url] message variable. This PR enables ChatAnthropic to
also accept image urls in said variable and makes it compatible with
OpenAI messages to make model switching easier.

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-04-18 15:15:45 -04:00
rylativity
dbf9986d44
langchain-ollama (partners) / langchain-core: allow passing ChatMessages to Ollama (including arbitrary roles) (#30411)
Replacement for PR #30191 (@ccurme)

**Description**: currently, ChatOllama [will raise a value error if a
ChatMessage is passed to
it](https://github.com/langchain-ai/langchain/blob/master/libs/partners/ollama/langchain_ollama/chat_models.py#L514),
as described
https://github.com/langchain-ai/langchain/pull/30147#issuecomment-2708932481.

Furthermore, ollama-python is removing the limitations on valid roles
that can be passed through chat messages to a model in ollama -
https://github.com/ollama/ollama-python/pull/462#event-16917810634.

This PR removes the role limitations imposed by langchain and enables
passing langchain ChatMessages with arbitrary 'role' values through the
langchain ChatOllama class to the underlying ollama-python Client.

As this PR relies on [merged but unreleased functionality in
ollama-python](
https://github.com/ollama/ollama-python/pull/462#event-16917810634), I
have temporarily pointed the ollama package source to the main branch of
the ollama-python github repo.

Format, lint, and tests of new functionality passing. Need to resolve
issue with recently added ChatOllama tests. (Now resolved)

**Issue**: resolves #30122 (related to ollama issue
https://github.com/ollama/ollama/issues/8955)

**Dependencies**: no new dependencies

[x] PR title
[x] PR message
[x] Lint and test: format, lint, and test all running successfully and
passing

---------

Co-authored-by: Ryan Stewart <ryanstewart@Ryans-MacBook-Pro.local>
Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-04-18 10:07:07 -04:00
ccurme
61d2dc011e
openai: release 0.3.14 (#30908) 2025-04-17 10:49:14 -04:00
ccurme
f0f90c4d88
anthropic: release 0.3.12 (#30907) 2025-04-17 14:45:12 +00:00
ccurme
add6a78f98
standard-tests, openai[patch]: add support standard audio inputs (#30904) 2025-04-17 10:30:57 -04:00
ccurme
86d51f6be6
multiple: permit optional fields on multimodal content blocks (#30887)
Instead of stuffing provider-specific fields in `metadata`, they can go
directly on the content block.
2025-04-17 12:48:46 +00:00
湛露先生
ff2930c119
partners: bug fix check_imports.py exit code. (#30897)
Signed-off-by: zhanluxianshen <zhanluxianshen@163.com>
2025-04-17 08:02:23 -04:00
ccurme
fa362189a1
docs: document OpenAI reasoning summaries (#30882) 2025-04-16 19:21:14 +00:00
ccurme
ca39680d2a
ollama: release 0.3.2 (#30865) 2025-04-16 09:14:57 -04:00
ccurme
085baef926
ollama[patch]: support standard image format (#30864)
Following https://github.com/langchain-ai/langchain/pull/30746
2025-04-15 22:14:50 +00:00
ccurme
47ded80b64
ollama[patch]: fix generation info (#30863)
https://github.com/langchain-ai/langchain/pull/30778 (not released)
broke all invocation modes of ChatOllama (intent was to remove
`"message"` from `generation_info`, but we turned `generation_info` into
`stream_resp["message"]`), resulting in validation errors.
2025-04-15 19:22:58 +00:00
Sydney Runkle
cf2697ec53
chroma: release 0.2.3 (#30860) 2025-04-15 14:11:23 -04:00
ccurme
8e9569cbc8
perplexity: release 0.1.1 (#30859) 2025-04-15 18:02:15 +00:00
ccurme
dd5f5902e3
openai: release 0.3.13 (#30858) 2025-04-15 17:58:12 +00:00
ccurme
3382ee8f57
anthropic: release 0.3.11 (#30857) 2025-04-15 17:57:00 +00:00
ccurme
9cfe6bcacd
multiple: multi-modal content blocks (#30746)
Introduces standard content block format for images, audio, and files.

## Examples

Image from url:
```
{
    "type": "image",
    "source_type": "url",
    "url": "https://path.to.image.png",
}
```


Image, in-line data:
```
{
    "type": "image",
    "source_type": "base64",
    "data": "<base64 string>",
    "mime_type": "image/png",
}
```


PDF, in-line data:
```
{
    "type": "file",
    "source_type": "base64",
    "data": "<base64 string>",
    "mime_type": "application/pdf",
}
```


File from ID:
```
{
    "type": "file",
    "source_type": "id",
    "id": "file-abc123",
}
```


Plain-text file:
```
{
    "type": "file",
    "source_type": "text",
    "text": "foo bar",
}
```
2025-04-15 09:48:06 -04:00
ccurme
f7c4965fb6
openai[patch]: update imports in test (#30828)
Quick fix to unblock CI, will need to address in core separately.
2025-04-14 19:33:38 +00:00
Marina Gómez
afd457d8e1
perplexity[patch]: Fix #30767: Handle missing citations attribute in ChatPerplexity (#30805)
This PR fixes an issue where ChatPerplexity would raise an
AttributeError when the citations attribute was missing from the model
response (e.g., when using offline models like r1-1776).

The fix checks for the presence of citations, images, and
related_questions before attempting to access them, avoiding crashes in
models that don't provide these fields.

Tested locally with models that omit citations, and the fix works as
expected.
2025-04-13 09:24:05 -04:00
ccurme
d9b628e764
xai: release 0.2.3 (#30790) 2025-04-11 14:05:11 +00:00
ccurme
9cfb95e621
xai[patch]: support reasoning content (#30758)
https://docs.x.ai/docs/guides/reasoning

```python
from langchain.chat_models import init_chat_model

llm = init_chat_model(
    "xai:grok-3-mini-beta",
    reasoning_effort="low"
)
response = llm.invoke("Hello, world!")
```
2025-04-11 14:00:27 +00:00
Sydney Runkle
8c6734325b
partners[lint]: run pyupgrade to get code in line with 3.9 standards (#30781)
Using `pyupgrade` to get all `partners` code up to 3.9 standards
(mostly, fixing old `typing` imports).
2025-04-11 07:18:44 -04:00
Jacob Lee
e72f3c26a0
fix(ollama): Remove redundant message from response_metadata (#30778) 2025-04-10 23:12:57 -07:00
célina
68361f9c2d
partners: (langchain-huggingface) Embeddings - Integrate Inference Providers and remove deprecated code (#30735)
Hi there, This is a complementary PR to #30733.
This PR introduces support for Hugging Face's serverless Inference
Providers (documentation
[here](https://huggingface.co/docs/inference-providers/index)), allowing
users to specify different providers

This PR also removes the usage of `InferenceClient.post()` method in
`HuggingFaceEndpointEmbeddings`, in favor of the task-specific
`feature_extraction` method. `InferenceClient.post()` is deprecated and
will be removed in `huggingface_hub` v0.31.0.

## Changes made

- bumped the minimum required version of the `huggingface_hub` package
to ensure compatibility with the latest API usage.
- added a provider field to `HuggingFaceEndpointEmbeddings`, enabling
users to select the inference provider.
- replaced the deprecated `InferenceClient.post()` call in
`HuggingFaceEndpointEmbeddings` with the task-specific
`feature_extraction` method for future-proofing, `post()` will be
removed in `huggingface-hub` v0.31.0.

 All changes are backward compatible.

---------

Co-authored-by: Lucain <lucainp@gmail.com>
Co-authored-by: ccurme <chester.curme@gmail.com>
2025-04-09 19:05:43 +00:00
Sydney Runkle
4556b81b1d
Clean up numpy dependencies and speed up 3.13 CI with numpy>=2.1.0 (#30714)
Generally, this PR is CI performance focused + aims to clean up some
dependencies at the same time.

1. Unpins upper bounds for `numpy` in all `pyproject.toml` files where
`numpy` is specified
2. Requires `numpy >= 2.1.0` for Python 3.13 and `numpy > v1.26.0` for
Python 3.12, plus a `numpy` min version bump for `chroma`
3. Speeds up CI by minutes - linting on Python 3.13, installing `numpy <
2.1.0` was taking [~3
minutes](https://github.com/langchain-ai/langchain/actions/runs/14316342925/job/40123305868?pr=30713),
now the entire env setup takes a few seconds
4. Deleted the `numpy` test dependency from partners where that was not
used, specifically `huggingface`, `voyageai`, `xai`, and `nomic`.

It's a bit unfortunate that `langchain-community` depends on `numpy`, we
might want to try to fix that in the future...

Closes https://github.com/langchain-ai/langchain/issues/26026
Fixes https://github.com/langchain-ai/langchain/issues/30555
2025-04-08 09:45:07 -04:00
湛露先生
9cbe91896e
Fix deepseek release tag, as it is update name. (#30717)
Thank you for contributing to LangChain!

- [ ] **PR title**: "package: description"
- Where "package" is whichever of langchain, community, core, etc. is
being modified. Use "docs: ..." for purely docs changes, "infra: ..."
for CI changes.
  - Example: "community: add foobar LLM"


- [ ] **PR message**: ***Delete this entire checklist*** and replace
with
    - **Description:** a description of the change
    - **Issue:** the issue # it fixes, if applicable
    - **Dependencies:** any dependencies required for this change
- **Twitter handle:** if your PR gets announced, and you'd like a
mention, we'll gladly shout you out!


- [ ] **Add tests and docs**: If you're adding a new integration, please
include
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in
`docs/docs/integrations` directory.


- [ ] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/

Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.

If no one reviews your PR within a few days, please @-mention one of
baskaryan, eyurtsev, ccurme, vbarda, hwchase17.

Signed-off-by: zhanluxianshen <zhanluxianshen@163.com>
2025-04-08 08:43:16 -04:00
ccurme
a2bec5f2e5
ollama: release 0.3.1 (#30716) 2025-04-07 20:31:25 +00:00
ccurme
e3f15f0a47
ollama[patch]: add model_name to response metadata (#30706)
Fixes [this standard
test](https://python.langchain.com/api_reference/standard_tests/integration_tests/langchain_tests.integration_tests.chat_models.ChatModelIntegrationTests.html#langchain_tests.integration_tests.chat_models.ChatModelIntegrationTests.test_usage_metadata).
2025-04-07 16:27:58 -04:00
ccurme
e106e9602f
groq[patch]: add retries to integration tests (#30707)
Tool-calling tests started intermittently failing with
> groq.APIError: Failed to call a function. Please adjust your prompt.
See 'failed_generation' for more details.
2025-04-07 12:45:53 -04:00
Tin Lai
4d03ba4686
langchain_qdrant: fix showing the missing sparse vector name (#30701)
**Description:** The error message was supposed to display the missing
vector name, but instead, it includes only the existing collection
configs.

This simple PR just includes the correct variable name, so that the user
knows the requested vector does not exist in the collection.

Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.

If no one reviews your PR within a few days, please @-mention one of
baskaryan, eyurtsev, ccurme, vbarda, hwchase17.

Signed-off-by: Tin Lai <tin@tinyiu.com>
2025-04-07 09:19:08 -04:00
Sydney Runkle
33ed7c31da
docs: fix perplexity install instructions in ChatPerplexity docstring (#30676)
* `openai` install no longer needs to be done manually
2025-04-04 12:58:18 -04:00
ccurme
59d508a2ee
openai[patch]: make computer test more reliable (#30672) 2025-04-04 13:53:59 +00:00
Sydney Runkle
17a9cd61e9
Bump langchain-core version in perplexity's pyproject.toml (#30647)
Blocking v0.1.0 release of `langchain-perplexity`
2025-04-03 16:19:10 +00:00
Sydney Runkle
3814bd1ea7
partners: Add Perplexity Chat Integration (#30618)
Perplexity's importance in the space has been growing, so we think it's
time to add an official integration!

Note: following the release of `langchain-perplexity` to `pypi`, we
should be able to add `perplexity` as an extra in
`libs/langchain/pyproject.toml`, but we're blocked by a circular import
for now.

---------

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-04-03 16:09:14 +00:00
ccurme
fe0fd9dd70
openai[patch]: upgrade tiktoken and fix test (#30621)
Related to https://github.com/langchain-ai/langchain/issues/30344

https://github.com/langchain-ai/langchain/pull/30542 introduced an
erroneous test for token counts for o-series models. tiktoken==0.8 does
not support o-series models in
`tiktoken.encoding_for_model(model_name)`, and this is the version of
tiktoken we had in the lock file. So we would default to `cl100k_base`
for o-series, which is the wrong encoding model. The test tested against
this wrong encoding (so it passed with tiktoken 0.8).

Here we update tiktoken to 0.9 in the lock file, and fix the expected
counts in the test. Verified that we are pulling
[o200k_base](https://github.com/openai/tiktoken/blob/main/tiktoken/model.py#L8),
as expected.
2025-04-02 10:44:48 -04:00
ccurme
816492e1d3
openai: release 0.3.12 (#30616) 2025-04-02 13:20:15 +00:00
Bagatur
111dd90a46
openai[patch]: support structured output and tools (#30581)
Co-authored-by: ccurme <chester.curme@gmail.com>
2025-04-02 09:14:02 -04:00
ccurme
8a69de5c24
openai[patch]: ignore file blocks when counting tokens (#30601)
OpenAI does not appear to document how it transforms PDF pages to
images, which determines how tokens are counted:
https://platform.openai.com/docs/guides/pdf-files?api-mode=chat#usage-considerations

Currently these block types raise ValueError inside
`get_num_tokens_from_messages`. Here we update to generate a warning and
continue.
2025-04-01 15:29:33 -04:00
Wenqi Li
64f97e707e
ollama[patch]: Support seed param for OllamaLLM (#30553)
**Description:** a description of the change
add the seed param for OllamaLLM client reproducibility

**Issue:** the issue # it fixes, if applicable
follow up of a similar issue
https://github.com/langchain-ai/langchain/issues/24703
see also https://github.com/langchain-ai/langchain/pull/24782

**Dependencies:** any dependencies required for this change
n/a
2025-03-31 11:28:49 -04:00
ccurme
b4fe1f1ec0
groq: release 0.3.2 (#30570) 2025-03-31 13:29:45 +00:00
Koshik Debanath
e7883d5b9f
langchain-openai: Support token counting for o-series models in ChatOpenAI (#30542)
Related to #30344

Add support for token counting for o-series models in
`test_token_counts.py`.

* **Update `_MODELS` and `_CHAT_MODELS` dictionaries**
- Add "o1", "o3", and "gpt-4o" to `_MODELS` and `_CHAT_MODELS`
dictionaries.

* **Update token counts**
  - Add token counts for "o1", "o3", and "gpt-4o" models.

---

For more details, open the [Copilot Workspace
session](https://copilot-workspace.githubnext.com/langchain-ai/langchain/pull/30542?shareId=ab208bf7-80a3-4b8d-80c4-2287486fedae).
2025-03-28 16:02:09 -04:00
omahs
6f8735592b
docs,langchain-community: Fix typos in docs and code (#30541)
Fix typos
2025-03-28 19:21:16 +00:00
Shixian Sheng
94a7fd2497
docs: fix broken hyperlinks in fireworks integration package README (#30538)
Fix two broken hyperlinks
2025-03-28 15:18:44 -04:00
ccurme
59908f04d4
fireworks: release 0.2.9 (#30527) 2025-03-27 16:04:20 +00:00
ccurme
05482877be
mistralai: release 0.2.10 (#30526) 2025-03-27 16:01:40 +00:00
Andras L Ferenczi
63673b765b
Fix: Enable max_retries Parameter in ChatMistralAI Class (#30448)
**partners: Enable max_retries in ChatMistralAI**

**Description**

- This pull request reactivates the retry logic in the
completion_with_retry method of the ChatMistralAI class, restoring the
intended functionality of the previously ineffective max_retries
parameter. New unit test that mocks failed/successful retry calls and an
integration test to confirm end-to-end functionality.

**Issue**
- Closes #30362

**Dependencies**
- No additional dependencies required

Co-authored-by: andrasfe <andrasf94@gmail.com>
2025-03-27 11:53:44 -04:00
ccurme
a9b1e1b177
openai: release 0.3.11 (#30503) 2025-03-26 19:24:37 +00:00
ccurme
8119a7bc5c
openai[patch]: support streaming token counts in AzureChatOpenAI (#30494)
When OpenAI originally released `stream_options` to enable token usage
during streaming, it was not supported in AzureOpenAI. It is now
supported.

Like the [OpenAI
SDK](f66d2e6fdc/src/openai/resources/completions.py (L68)),
ChatOpenAI does not return usage metadata during streaming by default
(which adds an extra chunk to the stream). The OpenAI SDK requires users
to pass `stream_options={"include_usage": True}`. ChatOpenAI implements
a convenience argument `stream_usage: Optional[bool]`, and an attribute
`stream_usage: bool = False`.

Here we extend this to AzureChatOpenAI by moving the `stream_usage`
attribute and `stream_usage` kwarg (on `_(a)stream`) from ChatOpenAI to
BaseChatOpenAI.

---

Additional consideration: we must be sensitive to the number of users
using BaseChatOpenAI to interact with other APIs that do not support the
`stream_options` parameter.

Suppose OpenAI in the future updates the default behavior to stream
token usage. Currently, BaseChatOpenAI only passes `stream_options` if
`stream_usage` is True, so there would be no way to disable this new
default behavior.

To address this, we could update the `stream_usage` attribute to
`Optional[bool] = None`, but this is technically a breaking change (as
currently values of False are not passed to the client). IMO: if / when
this change happens, we could accompany it with this update in a minor
bump.

--- 

Related previous PRs:
- https://github.com/langchain-ai/langchain/pull/22628
- https://github.com/langchain-ai/langchain/pull/22854
- https://github.com/langchain-ai/langchain/pull/23552

---------

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
2025-03-26 15:16:37 -04:00
ccurme
422ba4cde5
infra: handle flaky tests (#30501) 2025-03-26 13:28:56 -04:00
ccurme
299b222c53
mistral[patch]: check types in adding model_name to response_metadata (#30499) 2025-03-26 16:30:09 +00:00
ccurme
22d1a7d7b6
standard-tests[patch]: require model_name in response_metadata if returns_usage_metadata (#30497)
We are implementing a token-counting callback handler in
`langchain-core` that is intended to work with all chat models
supporting usage metadata. The callback will aggregate usage metadata by
model. This requires responses to include the model name in its
metadata.

To support this, if a model `returns_usage_metadata`, we check that it
includes a string model name in its `response_metadata` in the
`"model_name"` key.

More context: https://github.com/langchain-ai/langchain/pull/30487
2025-03-26 12:20:53 -04:00
ccurme
50ec4a1a4f
openai[patch]: attempt to make test less flaky (#30463) 2025-03-24 17:36:36 +00:00
ccurme
8486e0ae80
openai[patch]: bump openai sdk (#30461)
[New required
field](https://github.com/openai/openai-python/pull/2223/files#diff-530fd17eb1cc43440c82630df0ddd9b0893cf14b04065a95e6eef6cd2f766a44R26)
for `ResponseUsage` released in 1.66.5.
2025-03-24 12:10:00 -04:00
ccurme
cbbc968903
openai: release 0.3.10 (#30460) 2025-03-24 15:37:53 +00:00
ccurme
ed5e589191
openai[patch]: support multi-turn computer use (#30410)
Here we accept ToolMessages of the form
```python
ToolMessage(
    content=<representation of screenshot> (see below),
    tool_call_id="abc123",
    additional_kwargs={"type": "computer_call_output"},
)
```
and translate them to `computer_call_output` items for the Responses
API.

We also propagate `reasoning_content` items from AIMessages.

## Example

### Load screenshots
```python
import base64

def load_png_as_base64(file_path):
    with open(file_path, "rb") as image_file:
        encoded_string = base64.b64encode(image_file.read())
        return encoded_string.decode('utf-8')

screenshot_1_base64 = load_png_as_base64("/path/to/screenshot/of/application.png")
screenshot_2_base64 = load_png_as_base64("/path/to/screenshot/of/desktop.png")
```

### Initial message and response
```python
from langchain_core.messages import HumanMessage, ToolMessage
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="computer-use-preview",
    model_kwargs={"truncation": "auto"},
)

tool = {
    "type": "computer_use_preview",
    "display_width": 1024,
    "display_height": 768,
    "environment": "browser"
}
llm_with_tools = llm.bind_tools([tool])

input_message = HumanMessage(
    content=[
        {
            "type": "text",
            "text": (
                "Click the red X to close and reveal my Desktop. "
                "Proceed, no confirmation needed."
            )
        },
        {
            "type": "input_image",
            "image_url": f"data:image/png;base64,{screenshot_1_base64}",
        }
    ]
)

response = llm_with_tools.invoke(
    [input_message],
    reasoning={
        "generate_summary": "concise",
    },
)
response.additional_kwargs["tool_outputs"]
```

### Construct ToolMessage
```python
tool_call_id = response.additional_kwargs["tool_outputs"][0]["call_id"]

tool_message = ToolMessage(
    content=[
        {
            "type": "input_image",
            "image_url": f"data:image/png;base64,{screenshot_2_base64}"
        }
    ],
    #  content=f"data:image/png;base64,{screenshot_2_base64}",  # <-- also acceptable
    tool_call_id=tool_call_id,
    additional_kwargs={"type": "computer_call_output"},
)
```

### Invoke again
```python
messages = [
    input_message,
    response,
    tool_message,
]

response_2 = llm_with_tools.invoke(
    messages,
    reasoning={
        "generate_summary": "concise",
    },
)
```
2025-03-24 15:25:36 +00:00
Simon Paredes
df4448dfac
langchain-groq: Add response metadata when streaming (#30379)
- **Description:** Add missing `model_name` and `system_fingerprint`
metadata when streaming.

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-23 14:34:41 -04:00
ccurme
b78ae7817e
openai[patch]: trace strict in structured_output_kwargs (#30425) 2025-03-21 14:37:28 -04:00
ccurme
1de7fa8f3a
Revert "deepseek: temporarily bypass tests" (#30424)
Reverts langchain-ai/langchain#30423
2025-03-21 17:14:31 +00:00
ccurme
c74dfff836
deepseek: temporarily bypass tests (#30423)
Deepseek infra is not stable enough to get through integration tests.

Previous two attempts had two tests time out, they both pass locally.
2025-03-21 17:08:35 +00:00
ccurme
7147903724
deepseek: release 0.1.3 (#30422) 2025-03-21 16:39:50 +00:00
Andras L Ferenczi
b5f49df86a
partner: ChatDeepSeek on openrouter not returning reasoning (#30240)
Deepseek model does not return reasoning when hosted on openrouter
(Issue [30067](https://github.com/langchain-ai/langchain/issues/30067))

the following code did not return reasoning:

```python
llm = ChatDeepSeek( model = 'deepseek/deepseek-r1:nitro', api_base="https://openrouter.ai/api/v1", api_key=os.getenv("OPENROUTER_API_KEY")) 
messages = [
    {"role": "system", "content": "You are an assistant."},
    {"role": "user", "content": "9.11 and 9.8, which is greater? Explain the reasoning behind this decision."}
]
response = llm.invoke(messages, extra_body={"include_reasoning": True})
print(response.content)
print(f"REASONING: {response.additional_kwargs.get('reasoning_content', '')}")
print(response)
```

The fix is to extract reasoning from
response.choices[0].message["model_extra"] and from
choices[0].delta["reasoning"]. and place in response additional_kwargs.
Change is really just the addition of a couple one-sentence if
statements.

---------

Co-authored-by: andrasfe <andrasf94@gmail.com>
Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-21 16:35:37 +00:00
ccurme
e8e3b2bfae
ollama: release 0.3.0 (#30420) 2025-03-21 15:50:08 +00:00
Bob Merkus
5700646cc5
ollama: add reasoning model support (e.g. deepseek) (#29689)
# Description
This PR adds reasoning model support for `langchain-ollama` by
extracting reasoning token blocks, like those used in deepseek. It was
inspired by
[ollama-deep-researcher](https://github.com/langchain-ai/ollama-deep-researcher),
specifically the parsing of [thinking
blocks](6d1aaf2139/src/assistant/graph.py (L91)):
```python
  # TODO: This is a hack to remove the <think> tags w/ Deepseek models 
  # It appears very challenging to prompt them out of the responses 
  while "<think>" in running_summary and "</think>" in running_summary:
      start = running_summary.find("<think>")
      end = running_summary.find("</think>") + len("</think>")
      running_summary = running_summary[:start] + running_summary[end:]
```

This notes that it is very hard to remove the reasoning block from
prompting, but we actually want the model to reason in order to increase
model performance. This implementation extracts the thinking block, so
the client can still expect a proper message to be returned by
`ChatOllama` (and use the reasoning content separately when desired).

This implementation takes the same approach as
[ChatDeepseek](5d581ba22c/libs/partners/deepseek/langchain_deepseek/chat_models.py (L215)),
which adds the reasoning content to
chunk.additional_kwargs.reasoning_content;
```python
  if hasattr(response.choices[0].message, "reasoning_content"):  # type: ignore
      rtn.generations[0].message.additional_kwargs["reasoning_content"] = (
          response.choices[0].message.reasoning_content  # type: ignore
      )
```

This should probably be handled upstream in ollama + ollama-python, but
this seems like a reasonably effective solution. This is a standalone
example of what is happening;

```python
async def deepseek_message_astream(
    llm: BaseChatModel,
    messages: list[BaseMessage],
    config: RunnableConfig | None = None,
    *,
    model_target: str = "deepseek-r1",
    **kwargs: Any,
) -> AsyncIterator[BaseMessageChunk]:
    """Stream responses from Deepseek models, filtering out <think> tags.

    Args:
        llm: The language model to stream from
        messages: The messages to send to the model

    Yields:
        Filtered chunks from the model response
    """
    # check if the model is deepseek based
    if (llm.name and model_target not in llm.name) or (hasattr(llm, "model") and model_target not in llm.model):
        async for chunk in llm.astream(messages, config=config, **kwargs):
            yield chunk
        return

    # Yield with a buffer, upon completing the <think></think> tags, move them to the reasoning content and start over
    buffer = ""
    async for chunk in llm.astream(messages, config=config, **kwargs):
        # start or append
        if not buffer:
            buffer = chunk.content
        else:
            buffer += chunk.content if hasattr(chunk, "content") else chunk

        # Process buffer to remove <think> tags
        if "<think>" in buffer or "</think>" in buffer:
            if hasattr(chunk, "tool_calls") and chunk.tool_calls:
                raise NotImplementedError("tool calls during reasoning should be removed?")
            if "<think>" in chunk.content or "</think>" in chunk.content:
                continue
            chunk.additional_kwargs["reasoning_content"] = chunk.content
            chunk.content = ""
        # upon block completion, reset the buffer
        if "<think>" in buffer and "</think>" in buffer:
            buffer = ""
        yield chunk

```

# Issue
Integrating reasoning models (e.g. deepseek-r1) into existing LangChain
based workflows is hard due to the thinking blocks that are included in
the message contents. To avoid this, we could match the `ChatOllama`
integration with `ChatDeepseek` to return the reasoning content inside
`message.additional_arguments.reasoning_content` instead.

# Dependenices
None

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-21 15:44:54 +00:00
ccurme
d8145dda95
xai: release 0.2.2 (#30403) 2025-03-20 20:25:16 +00:00
ccurme
e194902994
mistral: release 0.2.9 (#30402) 2025-03-20 20:22:24 +00:00
ccurme
49466ec9ca
groq: release 0.3.1 (#30401) 2025-03-20 20:19:49 +00:00
ccurme
db1e340387
fireworks: release 0.2.8 (#30400) 2025-03-20 16:15:51 -04:00
ccurme
de3960d285
multiple: enforce standards on tool_choice (#30372)
- Test if models support forcing tool calls via `tool_choice`. If they
do, they should support
  - `"any"` to specify any tool
  - the tool name as a string to force calling a particular tool
- Add `tool_choice` to signature of `BaseChatModel.bind_tools` in core
- Deprecate `tool_choice_value` in standard tests in favor of a boolean
`has_tool_choice`

Will follow up with PRs in external repos (tested in AWS and Google
already).
2025-03-20 17:48:59 +00:00
ccurme
b86cd8270c
multiple: support strict and method in with_structured_output (#30385) 2025-03-20 13:17:07 -04:00
Mohammad Mohtashim
1103bdfaf1
(Ollama) Fix String Value parsing in _parse_arguments_from_tool_call (#30154)
- **Description:** Fix String Value parsing in
_parse_arguments_from_tool_call
- **Issue:** #30145

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-19 21:47:18 -04:00
ccurme
aae8306d6c
groq: release 0.3.0 (#30374) 2025-03-19 15:23:30 +00:00
Ashwin
83cfb9691f
Fix typo: change 'ben' to 'be' in comment (#30358)
**Description:**  
This PR fixes a minor typo in the comments within
`libs/partners/openai/langchain_openai/chat_models/base.py`. The word
"ben" has been corrected to "be" for clarity and professionalism.

**Issue:**  
N/A

**Dependencies:**  
None
2025-03-19 10:35:35 -04:00
Lance Martin
46d6bf0330
ollama[minor]: update default method for structured output (#30273)
From function calling to Ollama's [dedicated structured output
feature](https://ollama.com/blog/structured-outputs).

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-18 12:44:22 -04:00
ccurme
b91daf06eb
groq[minor]: remove default model (#30341)
The default model for `ChatGroq`, `"mixtral-8x7b-32768"`, is being
retired on March 20, 2025. Here we remove the default, such that model
names must be explicitly specified (being explicit is a good practice
here, and avoids the need for breaking changes down the line). This
change will be released in a minor version bump to 0.3.

This follows https://github.com/langchain-ai/langchain/pull/30161
(released in version 0.2.5), where we began generating warnings to this
effect.

![Screenshot 2025-03-18 at 10 33
27 AM](https://github.com/user-attachments/assets/f1e4b302-c62a-43b0-aa86-eaf9271e86cb)
2025-03-18 10:50:34 -04:00
ccurme
5684653775
openai[patch]: release 0.3.9 (#30325) 2025-03-17 16:08:41 +00:00
ccurme
eb9b992aa6
openai[patch]: support additional Responses API features (#30322)
- Include response headers
- Max tokens
- Reasoning effort
- Fix bug with structured output / strict
- Fix bug with simultaneous tool calling + structured output
2025-03-17 12:02:21 -04:00
ccurme
c74e7b997d
openai[patch]: support structured output via Responses API (#30265)
Also runs all standard tests using Responses API.
2025-03-14 15:14:23 -04:00