langchain-anthropic
An adapter package in the LangChain ecosystem for connecting to Anthropic Claude models. It provides a unified invocation interface and capability description within LangChain. The package is in a period of stable iteration, tracking new Claude model releases and features while exposing them through LangChain's standard chat model abstractions.
Timeline (last 90 days)
If it is only routine usage and does not yet involve changing instructions mid-session, langchain-anthropic 1.7.6 can be upgraded at a normal pace when convenient, without disrupting the established schedule.
langchain-anthropic 1.7.6 upgrades urllib3 from 2.7.0 to 2.8.0.
The changes from 1.7.5 to 1.7.6 mainly fall into three categories of work: model configuration data refreshes, cross-vendor capability declarations, and dependency version upgrades.
langchain-anthropic 1.7.6 begins declaring "mid-conversation" support capability in the model profiles of the core library, Anthropic and OpenAI.
This declaration of mid-conversation support involves three packages at the same time: core, anthropic and openai
If a project has langchain-core, langchain-openai, langchain-anthropic and other packages installed at the same time, it is recommended to upgrade them all to mutually compatible versions to avoid inconsistency between capability declarations and actual implementations.
The changes in PR #41180 involve three packages: core, anthropic and openai.
The previous version before 1.7.5 is 1.7.4.
1.7.5 fixes an issue where invalid tool calls were serialized as tool use during replay (PR #40864).
langchain-anthropic 1.7.5 adds compatibility support for Claude Sonnet 5.5 (PR #40882).
2 more timeline events
All Facts (6)
LangChain's Claude integration is implemented through the langchain-anthropic sub-package, but the foundational abstractions for model compatibility are defined in langchain-core.
70%Unverifiedlangchain-anthropic 1.7.6 begins declaring "mid-conversation" support capability in the model profiles of the core library, Anthropic and OpenAI.
50%UnverifiedIf a project has langchain-core, langchain-openai, langchain-anthropic and other packages installed at the same time, it is recommended to upgrade them all to mutually compatible versions to avoid inconsistency between capability declarations and actual implementations.
50%Unverifiedlangchain-anthropic 1.7.5 adds compatibility support for Claude Sonnet 5.5 (PR #40882).
50%Unverified1.7.5 fixes an issue where invalid tool calls were serialized as tool use during replay (PR #40864).
50%UnverifiedSome integrations with higher usage frequency or official partnership relationships (such as OpenAI, Anthropic, and Google) have been split into independent partner packages, for example langchain-openai and langchain-anthropic.
50%