TL;DR

LangChain.js now correctly handles and emits reasoning blocks from AWS Bedrock, ensuring compatibility with both new and existing workflows.

Fixes Worth Knowing

• Fixed an issue where string-based AI message content (common in LangGraph) could cause errors.
• Resolved a problem with malformed tool input schemas that previously caused unrecoverable errors.
• OpenAI responses now consistently include the output_text field for assistant content.
• Improved handling of Bedrock reasoning blocks, preserving signatures and compatibility.

TL;DR

This release updates integrations for several LLM providers (large language models) to align with the latest @langchain/[email protected] version, ensuring compatibility and access to recent OpenAI improvements.

Fixes Worth Knowing

Integrations with Fireworks, OpenRouter, Together AI, and Neo4j have been updated to maintain compatibility with the core LangChain.js library and OpenAI.

TL;DR

Gemini models now support native streaming events, improving the responsiveness of applications using the @langchain/google-genai and @langchain/google-common packages.

New

• Streaming events are now natively supported for Gemini models, providing a more fluid user experience (streaming data).
• Gemini API requests now automatically include necessary configuration for mixed server-side tools and function declarations, preventing errors.

Fixes Worth Knowing

• A crash in @langchain/google-common when processing tool messages with plain object content has been resolved.

TL;DR

LangChain.js now supports ToolRuntime (a way to define tools), improving tool integration and structured output parsing, especially with Ollama and OpenAI models.

Fixes Worth Knowing

• Message ordering after summarization is now correct.
• PDF loading now explicitly requires pdf-parse v1 (a PDF parsing library).
• Tool calls from Ollama models now correctly parse structured outputs.
• Chat models are now compatible with Vercel deployments (a web hosting platform).

Before You Upgrade

Ensure you are using pdf-parse v1 if you utilize the PDF loader.

TL;DR

LangChain.js now features more granular entry points (specific starting points) for increased modularity and flexibility.

New

LangChain.js now offers more granular entry points, allowing developers to import only the components they need, reducing bundle size and improving performance.

Fixes Worth Knowing

Prisma documentation has been updated for improved clarity and usability (database toolkit).