Custom authentication & access control for LangGraph Platform
Today, we're thrilled to announce Custom Authentication and Resource-Level Access Control for Python deployments in LangGraph Cloud and self-hosted...
🔁 Modify graph state from tools in LangGraph
LangGraph's latest update gives you greater control over your agents by enabling tools to directly update the graph state. This is a game-changer for use...
🔍 Diff view in LangSmith's prompt hub
We’ve added a new Diff View feature in LangSmith’s Prompt Hub. As prompts evolve, keeping track of changes can get tricky—especially when iterating rapidly....
`interrupt`: Simplifying human-in-the-loop agents
Our latest feature in LangGraph, interrupt , makes building human-in-the-loop workflows easier. Why Human-in-the-Loop? Agents aren’t perfect, so keeping...
File attachments with example datasets in LangSmith
LangSmith now supports uploading any binary file (e.g. images, audio, videos, PDF) directly with examples in your datasets. You can now run multimodal LLM...
🤖 Command in LangGraph to build edgeless graphs & multi-agent workflows
We've released Command , a new tool in LangGraph that lets you manage dynamic, edgeless agent flows. Now, nodes can dynamically decide which node to execute...
📈 Organization Usage Charts for Self-Hosted Customers
Self-hosted LangSmith customers can now access organization usage charts . Track trace usage across your entire org or by workspace with ease. What’s New:...
🔭 OpenTelemetry support for LangSmith
LangSmith now supports OpenTelemetry, bringing distributed tracing and end-to-end visibility to your LLM observability workflow. Ingest traces in OpenLLMetry...
🛠️ LangSmith SDK v0.2: Simplified evaluators and improved performance
We've recently released v0.2 of our LangSmith SDKs (Python and TypeScript) with a number of quality-of-life improvements to make the evals experience more...
🔍 Semantic search for LangGraph's long-term memory
With semantic search for LangGraph’s long-term memory, your agents can now find relevant memories based on meaning , not just exact matches — making it...