LangGraph is a multi-agent framework. This means not only interacting with other LangGraph agents, but all other types of agents as well, regardless of how they are built. Today we are taking a few steps to build towards this vision. We are announcing: Agent Protocol: a common interface for agent

LangSmith's homepage is now organized into Observability, Evaluation, and Prompt Engineering. Learn why we organized the homepage like this. Plus, see our latest Resource Tags updates.

Learn how Dun & Bradstreet, a leading financial data analytics company, empowers global clients with business data — from credit risk to ownership structures — to make better decisions using their AI assistant, Chat D&B. See how LangChain and LangSmith have helped Dun & Bradstreet on their journey.

Promptim is an experimental prompt optimization library to help you systematically improve your AI systems. Promptim automates the process of improving prompts on specific tasks. You provide initial prompt, a dataset, and custom evaluators (and optional human feedback), and promptim runs an optimization loop to produce

Use Prompt Canvas in LangSmith to collaborate with an AI agent to build and optimize your prompts.

We are excited to launch SWE-Kit, an open-source headless IDE with AI-native coding toolkits for AI agents, as part of Composio's agent tooling ecosystem. SWE-kit offers a headless IDE featuring Language Server Protocol (LSP) for code intelligence and a development container for secure code execution. It also features

Editor's Note: we're EXTREMELY excited to highlight this research from Ankush Garg and Shreya Shankar from Berkeley. At LangChain, two of the biggest problems we think about are evals and agents, and this research sits right at the intersection. You can try it out today in

Editor's Note: one of the most common use cases we've seen for LangGraph is complex research agents. This guest blog post by Chaos Labs highlights a great example of that. It utilizes multiple sources and complex architecture to do research that would power resolution in prediction

We've rebranded our service for deploying and scaling LangGraph apps as LangGraph Platform. Learn about the multiple deployment options and what LangGraph Platform entails.

“What we’ve got here is failure to communicate” - Cool Hand Luke (1967) Communication is the hardest part of life. It’s also the hardest part of building LLM applications. New hires always requires a lot of communication when first joining a company, no matter

Reflections on how LangChain has evolved — including our products, ecosystem, and community — over the past two years, and where we're headed next.

At Sequoia’s AI Ascent conference in March, I talked about three limitations for agents: planning, UX, and memory. Check out that talk here. In this post I will dive more into memory. See the previous post on planning here, and the previous posts on UX here, here, and

See how Rexera migrated to LangGraph to create a robust quality control agent for real estate workflows, significantly improving their LLM response accuracy.

This is a guest blog post written by Sam and Connor at Unify. Unify is reinventing how go-to-market teams work using generative AI. As part of this transformation, they are launching a new agents feature today (powered by LangGraph and LangSmith). We had the pleasure of learning more about the

Today, we are excited to announce the first steps towards long-term memory support in LangGraph, available both in Python and JavaScript. Long-term memory lets you store and recall information between conversations so your agent can learn from feedback and adapt to user preferences. This feature is part of the OSS