langchain-ai/langchain

langchain

Performance History

Latest Results

feat(core,anthropic,openai): declare mid-conversation support in model profiles (#41180) Alternative to #41175 (#41150). `ChatAnthropic` decides whether to keep a mid-conversation `SystemMessage` in place by matching model names. That misses Bedrock model IDs, and it makes callers such as deepagents keep their own model and class allowlists. This PR moves the decision into the model profile. - `ModelProfile` gets two fields, `mid_conversation_system_messages` and `mid_conversation_tools`. The second covers adding a tool by full definition or by reference. The block format stays provider-specific. - `ChatAnthropic` reads `mid_conversation_system_messages` from its profile instead of a list of model names. - A chat model whose API can't send a capability turns it off in `_resolve_model_profile`. `ChatOpenAI` does this when it isn't on the Responses API, and `_ChatOpenAICodex` does it for both fields. `AzureChatOpenAI` makes no claim, because the live API tests didn't cover Azure. - The profile data comes from the live API tests in #41175 and langchain-ai/deepagents#6874. A caller then checks one field: ```python if (model.profile or {}).get("mid_conversation_tools"): ... # add the tool in a message ``` ## Review notes - Bedrock still needs the same two fields in langchain-aws's profile data, in a follow-up PR there. - A new model ID now needs a profile entry. The old prefix list matched new releases automatically. - Passing `profile=` replaces the resolved profile, so it drops these flags, as it already drops `reasoning_effort_levels`. - The partners need a langchain-core release with the new fields first. Otherwise they warn about unknown profile keys. ## Release note `ModelProfile` gains `mid_conversation_system_messages` and `mid_conversation_tools`. `ChatAnthropic` now decides whether to keep a mid-conversation `SystemMessage` in place from its profile, not its model name. Claude Sonnet 5 and Haiku 5.5 now keep it in place. Claude Haiku 5.5 also gets a profile, so its default `max_tokens` rises from 4096 to 128000. _Written with the help of an AI coding agent._

master

11 hours ago

refactor(core,anthropic,openai): rename `mid_conversation_tool_definitions` to `mid_conversation_tools` One flag covers both kinds of mid-conversation tool change: a full `tool_definition` and a `tool_reference` by name. Providers differ in the forms they accept (OpenAI takes definitions only, Anthropic also takes references and removals), but callers already pick the block by provider, so a second flag would only matter once LangChain has a standard block.

richardscarrott/core/mid-conversation-profile-fields

12 hours ago

docs(core): mid_conversation_tool_definitions says nothing about references

richardscarrott/core/mid-conversation-profile-fields

14 hours ago

refactor(core,anthropic,openai): rename `inline_tool_definitions` to `mid_conversation_tool_definitions` Match Anthropic's "mid-conversation tool changes" and the sibling `mid_conversation_system_messages` field. In Anthropic's beta naming "inline" means "by value", which would make a future references field read as `inline_tool_references`. The docstring now says the field covers definitions only.

richardscarrott/core/mid-conversation-profile-fields

14 hours ago

feat(core,anthropic,openai): declare mid-conversation support in model profiles Add `mid_conversation_system_messages` and `inline_tool_definitions` to `ModelProfile`. `ChatAnthropic` reads the first from its profile instead of matching model names. `ChatOpenAI` off the Responses API and `_ChatOpenAICodex` turn the capabilities off in `_resolve_model_profile`, and `AzureChatOpenAI` makes no claim. Alternative to #41175.

richardscarrott/core/mid-conversation-profile-fields

15 hours ago

Merge branch 'master' into uditstocks/langchain/jump-to-tools-pending-dispatch

uditstocks:uditstocks/langchain/jump-to-tools-pending-dispatch

17 hours ago

Merge branch 'master' into core/format-context

saptarshiroy39:core/format-context

1 day ago

fix(langchain): stop rejecting edit args the tool doesn't declare

fix/hitl-per-call-interrupts

1 day ago

Latest Branches

feat(core,anthropic,openai): declare mid-conversation support in model profiles#41180

12 hours ago

370146a

richardscarrott/core/mid-conversation-profile-fields

fix(langchain): stop `jump_to="tools"` from bypassing HITL review#41023

17 hours ago

6e9d7c6

uditstocks:uditstocks/langchain/jump-to-tools-pending-dispatch

feat(core): add format_context helper for documents#41168

1 day ago

9a26c64

saptarshiroy39:core/format-context

Home Terms Privacy Docs