π₯ Dynamic few-shot example selection in open beta
Dynamic few-shot example selection is now available in open beta for LangSmith users (currently for Cloud only). With this feature, you can perform...
π LangSmith Playground v2: More models, better compatibility, faster testing
Thousands of engineers use the LangSmith Playground as their IDE for prompt engineering. Now, weβve made the Playground even better β hereβs whatβs new:...
πͺ Generate synthetic examples to enhance a LangSmith dataset
You can now generate synthetic examples for your datasets in LangSmith. Using the dataset schema you've defined, we generate new examples based on existing...
βοΈ π€ LangGraph Cloud is available in open beta
LangGraph Cloud is now available in open beta. LangGraph Cloud lets you build fault-tolerant, scalable agents. You can: All LangSmith users on Plus and...
βοΈ LangGraph v0.2: Increased customization with new checkpointers
We've released LangGraph v0.2 for increased customization with new checkpointer libraries. LangGraph v0.2 allows you to tailor stateful LangGraph apps to...
π Few-shot examples with LangSmith datasets
LangSmith now supports fast, easy-to-use APIs to find few shot examples from your datasets, which you can use to improve your LLM applications without...
π LangGraph Studio: The first agent IDE
LangGraph Studio offers a new way to develop LLM applications by providing a specialized agent IDE that enables visualization, interaction, and debugging of...
π Trace comparison view & saving custom trace filters
In the runs table in LangSmith, you can now select traces to compare side-by-side. This lets you make direct comparisons to see discrepancies in output...
π Define, validate, and update dataset schemas in LangSmith
When building a dataset iteratively for an LLM app, having a defined schema for testing across examples lets you avoid broken code and keep your data clean...
βοΈ Multiple annotators can review a run in LangSmith
LangSmith's annotation queue now supports allowing multiple people to review an individual run. This makes it easier to coordinate data review across...