Anthropic's API rejects `tool_use`/`tool_result` IDs that don't match
`^[a-zA-Z0-9_-]+$` with a 400. When a conversation thread is replayed
across providers — e.g. a user switches a running thread from Kimi (via
Fireworks) to Claude — the prior turns carry tool-call IDs minted by the
other provider (Kimi emits `functions.write_todos:0`, whose `.` and `:`
are invalid), and the request fails. Tool-call IDs are now normalized to
an Anthropic-compatible form during request formatting, with the
original `tool_use.id` and its paired `tool_use_id` mapped identically
so they stay linked.
Sweep classic deprecations so every removal lands on `2.0.0`, runtime
warnings carry the auto-generated since/removal/alternative line, and
replacements steer at `langchain.agents.create_agent` and
`with_structured_output(...)` instead of pre-v1 LangGraph +
`python.langchain.com` links.
## Changes
- **Bump removal targets from `1.0` / `1.0.0` to `2.0.0`** across
agents, chains, memory, retrievers, structured-output, vectorstore
toolkits, and the `langchain_classic._api.module_import` shim — gives
users a real runway now that v1 has shipped.
- **Move bespoke `message=` strings onto `addendum=`** (or split into
`alternative=` + `addendum=`). `warn_deprecated` skips the
auto-generated since/removal/alternative line whenever `message=` is
set, so the prior pattern silently dropped that info from the runtime
`LangChainDeprecationWarning`. Matches the pattern already used in
`HTMLHeaderTextSplitter.split_text_from_url`, which is updated for
consistency.
- **Repoint `alternative=` at v1 replacements**: chains/memory/agent
toolkits → `langchain.agents.create_agent` (with checkpointer or
retrieval-tool guidance in the addendum); `openai_functions` and
`chains/structured_output` → `ChatModel.with_structured_output(...)`;
`openapi` chains → `ChatModel.bind_tools(...)` + HTTP client.
`ConversationChain` no longer points at `RunnableWithMessageHistory`.
- **Refresh `AGENT_DEPRECATION_WARNING`** in
`langchain_classic._api.deprecation` — drop stale LangGraph and
`python.langchain.com` links in favor of `langchain.agents.create_agent`
and the `docs.langchain.com/oss/python/migrate/langchain-v1` guide.
Propagates to all 13 caller sites in `agents/`.
- **Newly deprecate `langchain_classic.chat_models.init_chat_model` and
`langchain_classic.embeddings.init_embeddings`** with the framing
*"maintained in `langchain`; `langchain-classic` retains this entry
point for import-compatibility only"*. The classic docstring examples
and the warning admonition both point at `langchain.chat_models`.
- **Improve `init_chat_model` docstrings** in both `langchain_v1` and
the classic copy: clarify `provider:model` prefix vs. `model_provider=`,
recommend pinned IDs over moving aliases, add the `upstage` provider
row, and refresh examples to GA models (`gpt-5.5`, `claude-opus-4-7`).
- **Standardize partner Anthropic deprecations**: replace
`AnthropicLLM`'s `model_validator(raise_warning)` with
`@deprecated(since="0.1.0", removal="2.0.0",
alternative="ChatAnthropic")`, and pin the `ChatAnthropic`
`output_format` runtime warning at `langchain-anthropic 2.0.0` instead
of "a future version".
Closes#37042
---
`AnthropicPromptCachingMiddleware` was unconditionally setting top-level
`cache_control` in `model_settings` for any `ChatAnthropic` subclass.
That field is direct-Anthropic-API only — `ChatAnthropicBedrock` (which
subclasses `ChatAnthropic` and passed the existing `isinstance` gate)
errored with `cache_control: Extra inputs are not permitted`.
Investigating that surfaced a related regression: PR #35967 also deleted
the block-level `cache_control` injection in `_get_request_payload`,
which silently disabled caching entirely for non-direct subclasses
(Bedrock had been falling back to in-block breakpoints). This restores
both paths.
## Changes
- Add `_is_direct_anthropic_llm_type` predicate that allowlists
`_llm_type == "anthropic-chat"`. Both the middleware's
`_supports_automatic_caching` and the new branch in
`ChatAnthropic._get_request_payload` route through it, so any subclass
that overrides `_llm_type` (Bedrock today, future direct-API variants
tomorrow) is treated as non-direct by default. Replaces the prior
substring-matching denylist on `"bedrock"`/`"vertex"`.
- Restore `_collect_code_execution_tool_ids`,
`_is_code_execution_related_block`, and a new
`_apply_cache_control_to_last_eligible_block` helper in `chat_models`.
For non-direct subclasses, `_get_request_payload` now pops
`cache_control` from kwargs and walks messages newest-to-oldest,
attaching the breakpoint to the last block that isn't
`code_execution`-related (Anthropic forbids breakpoints on those).
- Emit `UserWarning` when `cache_control` is requested but every
candidate block is `code_execution`-related — previously a silent drop.
- `AnthropicPromptCachingMiddleware._apply_caching` now sets the
top-level `cache_control` only when
`_supports_automatic_caching(request.model)`. System-message and
tool-definition breakpoints continue to apply for all `ChatAnthropic`
subclasses, since those are accepted by every transport.
- Note: `ChatAnthropicVertex` does not subclass `ChatAnthropic` (it
lives in `langchain-google-vertexai` and ships its own
`_get_request_payload`), so the chat-models changes here only affect
Bedrock. The middleware-side gate covers Vertex implicitly via the
`isinstance(request.model, ChatAnthropic)` check that already excludes
it.
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
Prevent double counting, since the sum of all `input_token_details`
should never exceed `input_tokens`
---------
Co-authored-by: Chester Curme <chester.curme@gmail.com>
- use latest models in examples to highlight support
- standardize on using IDs in examples - no more aliases to improve
determinism in future tests
- bump lock
- in integration tests, fix stale casettes and use `MODEL_NAME`
uniformly where possible
- add case for default max tokens for sonnet-4-5 (was missing)
