langchain
https://github.com/langchain-ai/langchain
Python
⚡ Building applications with LLMs through composability ⚡
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported41 Subscribers
Add a CodeTriage badge to langchain
Help out
- Issues
- Tool `config: RunnableConfig` parameter receives the parent config instead of the child config, so nested runnables are traced as siblings of the tool run
- mistralai: an error event inside a chat stream is silently dropped, and the truncated output is returned as a successful response
- create_agent ends mixed batches early when a non-direct tool is registered at runtime
- `TextSplitter._merge_splits` does not warn when the last chunk exceeds `chunk_size`
- _generate_response_from_error leaves an unawaited coroutine when the error's response comes from an async HTTP client
- core/openai: with `output_version="v1"` and streaming, a tool call with invalid JSON arguments stays a raw `tool_call_chunk` block in `AIMessage.content` and is sent back to Chat Completions
- feat(langchain): add agent-visible model-call budgets
- `HuggingFacePipeline.from_model_id` crashes on device strings and misplaces models on non-CUDA accelerators (e.g. Intel XPU)
- langchain-core: External cancellation of BaseChatModel.ainvoke skips terminal callbacks
- core: InMemoryVectorStore returns arbitrary documents for negative k
- Docs
- Python not yet supported