Automated model-profile refresh PRs (e.g. #38210) ship a static template
body, so a reviewer has to open *Files changed* and read large blocks of
generated data to learn what actually moved. Because the underlying
profile data is fully structured, we can describe the changes
deterministically — no LLM, no hallucination risk.
This adds a `langchain-profiles summarize` subcommand that compares the
working-tree `_profiles.py` files against a git ref and renders a
skimmable Markdown summary: models added (with a short capability
descriptor), models removed, and per-field capability changes
(context/output tokens, modalities, tool calling, reasoning, etc.),
grouped by provider and capped so huge refreshes stay readable. Profiles
are read with `ast.literal_eval` rather than imported, so the generated
data file is never executed.
Example output for a refresh that adds a model and bumps an output
limit:
```
## Summary of changes
**1 added · 0 removed · 1 changed** across 1 provider(s).
### openai
**➕ 1 added**
- `gpt-6-preview` — 1,000,000 ctx, 128,000 out, text+image+audio in, reasoning, tools
**✏️ 1 changed**
- `gpt-3.5-turbo`: max output tokens 4,096 → 16,384
```
Made by [Open
SWE](https://openswe.vercel.app/agents/9bcbf182-effc-ba9b-0df3-afac620ad152)
---------
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
Originally a narrow bump of mypy to `1.20` in four packages. Expanded to
get the whole monorepo onto a single, current mypy and a consistent
type-check configuration, so contributors no longer hit different mypy
versions and divergent behavior depending on which package they touch.
### What changed
- **Unified the mypy pin to `>=2.1.0,<2.2.0`** in every mypy-using
package (6 libs + 14 partners), replacing the previously scattered pins
(`1.10`/`1.17`/`1.18`/`1.19`/`1.20`, with assorted upper bounds).
- **Unified the `[tool.mypy]` base per tier:**
- libs: `plugins = ["pydantic.mypy"]`, `strict = true`,
`enable_error_code = "deprecated"`, `warn_unreachable = true`
- partners: `disallow_untyped_defs = true`
- Normalized style (`disallow_untyped_defs = "True"` string → bool,
quote/key consistency).
- **Fixed the 20 real errors** mypy 2.1 surfaces: `redundant-cast` from
improved narrowing (`core`, `langchain-classic`), a `var-annotated` for
`_LOGGED`, a return-type widening in `langchain-groq`'s
`_convert_from_v1_to_groq` (it can legitimately return a bare `str`),
and stale `type-arg`/`unused-ignore` in `langchain-model-profiles`
tests.
### Deliberate non-uniformity (documented inline in the relevant
`pyproject.toml`s)
Going fully byte-identical would surface ~196 additional errors that are
*not* real bugs, so two settings are kept package-appropriate:
- **`warn_unreachable`** is enabled on every strict lib **except
`core`**, where it false-flags intentional defensive code — including
the SSRF / IP-policy guards in `_security/` — as unreachable.
- **`pydantic.mypy` plugin** is used only on `anthropic` and
`perplexity` (their code is authored against it and reports ~99/~132
errors without it). It is *not* added to the other partners, where it
only flags the public alias constructor API (e.g. `ChatGroq(model=...)`)
in tests rather than finding bugs.
- **`ollama`** is left on its `ty` type checker; it does not use mypy.
---------
Co-authored-by: Mason Daugherty <github@mdrxy.com>
As a LangChain user streaming a tool-calling model, I expect each
streamed chunk to expose structured `tool_call_chunk` content blocks so
I can render or process tool calls live, instead of waiting for the
final aggregated message.
This adds `tool_call_streaming` to `ModelProfile` and uses it in the
standard chat-model tool-calling tests. When a model profile opts in,
`test_tool_calling` and `test_tool_calling_async` now validate that at
least one streamed chunk includes a `tool_call_chunk` block via
`content_blocks`, while preserving the existing final-message
validation.
This keeps the contract profile-gated so providers can opt in once their
streaming chunk shape is verified. This PR opts in the providers
verified by smoke testing with straightforward profile coverage: OpenAI,
Anthropic, Fireworks, HuggingFace, OpenRouter, DeepSeek, and xAI. The
generated profile artifacts are refreshed so runtime profiles expose the
new capability flag.
Perplexity Responses also passed the smoke test, but its current profile
data is for the `sonar` family while the Responses smoke path used a
routed model string. That profile strategy is left as follow-up.
MistralAI currently streams `.tool_call_chunks`, but its content-block
translator exposes a complete `tool_call` block instead of
`tool_call_chunk`, so it also stays out of this flag until that
integration is fixed.
PR #35788 added 7 new fields to the `langchain-profiles` CLI output
(`name`, `status`, `release_date`, `last_updated`, `open_weights`,
`attachment`, `temperature`) but didn't update `ModelProfile` in
`langchain-core`. Partner packages like `langchain-aws` that set
`extra="forbid"` on their Pydantic models hit `extra_forbidden`
validation errors when Pydantic encountered undeclared TypedDict keys at
construction time. This adds the missing fields, makes `ModelProfile`
forward-compatible, provides a base-class hook so partners can stop
duplicating model-profile validator boilerplate, migrates all in-repo
partners to the new hook, and adds runtime + CI-time warnings for schema
drift.
## Changes
### `langchain-core`
- Add `__pydantic_config__ = ConfigDict(extra="allow")` to
`ModelProfile` so unknown profile keys pass Pydantic validation even on
models with `extra="forbid"` — forward-compatibility for when the CLI
schema evolves ahead of core
- Declare the 7 missing fields on `ModelProfile`: `name`, `status`,
`release_date`, `last_updated`, `open_weights` (metadata) and
`attachment`, `temperature` (capabilities)
- Add `_warn_unknown_profile_keys()` in `model_profile.py` — emits a
`UserWarning` when a profile dict contains keys not in `ModelProfile`,
suggesting a core upgrade. Wrapped in a bare `except` so introspection
failures never crash model construction
- Add `BaseChatModel._resolve_model_profile()` hook that returns `None`
by default. Partners can override this single method instead of
redefining the full `_set_model_profile` validator — the base validator
calls it automatically
- Add `BaseChatModel._check_profile_keys` as a separate
`model_validator` that calls `_warn_unknown_profile_keys`. Uses a
distinct method name so partner overrides of `_set_model_profile` don't
inadvertently suppress the check
### `langchain-profiles` CLI
- Add `_warn_undeclared_profile_keys()` to the CLI (`cli.py`), called
after merging augmentations in `refresh()` — warns at profile-generation
time (not just runtime) when emitted keys aren't declared in
`ModelProfile`. Gracefully skips if `langchain-core` isn't installed
- Add guard test
`test_model_data_to_profile_keys_subset_of_model_profile` in
model-profiles — feeds a fully-populated model dict to
`_model_data_to_profile()` and asserts every emitted key exists in
`ModelProfile.__annotations__`. CI fails before any release if someone
adds a CLI field without updating the TypedDict
### Partner packages
- Migrate all 10 in-repo partners to the `_resolve_model_profile()`
hook, replacing duplicated `@model_validator` / `_set_model_profile`
overrides: anthropic, deepseek, fireworks, groq, huggingface, mistralai,
openai (base + azure), openrouter, perplexity, xai
- Anthropic retains custom logic (context-1m beta → `max_input_tokens`
override); all others reduce to a one-liner
- Add `pr_lint.yml` scope for the new `model-profiles` package
Extract additional fields from models.dev into `_model_data_to_profile`:
`name`, `status`, `release_date`, `last_updated`, `open_weights`,
`attachment`, `temperature`
Move the model profile refresh logic from an inline bash script in the
GitHub Actions workflow into a `make refresh-profiles` target in
`libs/model-profiles/Makefile`. This makes it runnable locally with a
single command and keeps the provider map in one place instead of
duplicated between CI and developer docs.
- Sort model profiles alphabetically by model ID (the top-level
`_PROFILES` dictionary keys, e.g. `claude-3-5-haiku-20241022`,
`gpt-4o-mini`) before writing `_profiles.py`, so that regenerating
profiles only shows actual data changes in diffs — not random reordering
from the models.dev API response order
- Regenerate all 10 partner profile files with the new sorted ordering
The trailing comma regex in the profile generation script consumed the
closing `}` as part of its match, preventing nested closing braces from
getting their own trailing comma. This caused `ruff format` failures on
every generated `_profiles.py` file.
Switches to a lookahead (`(?=...)`) so the closing bracket is asserted
but not consumed, allowing each nesting level to independently receive
its trailing comma.
Fixes#35332.
- Add `text_inputs` and `text_outputs` fields to `ModelProfile`
- Regenerate `_profiles.py` for all providers
## Why
models.dev data includes `'text'` as both an input and output modality,
but we didn't capture it.
models.dev broadly contains models without text input (Whisper/ASR) and
without text output (image generators, TTS).
Without this, downstream consumers can't filter on model text support
(e.g. preventing users from passing text input to an audio-only model).
---
We'd need to also run for Google, AWS and cut releases for all to
propagate
