LangChain launches new evaluation features for LangSmith

LangChain Blog

LangSmith adds reusable evaluators with 30+ templates, public benchmark sharing, and Align Evals to calibrate LLM-as-a-judge. A Q&A benchmark over LangChain docs is the first shared dataset.

LangChainEvaluation

See 7 more sources

Aligning LLM-as-a-Judge with Human Preferences2 days ago

Reusable Evaluators and Evaluator Templates in LangSmith1 day ago

Testing Fine Tuned Open Source Models in LangSmith1 day ago

Sharing LangSmith Benchmarks1 day ago

Test Run Comparisons2 days ago

Regression Testing with LangSmith2 days ago

Introducing Align Evals: Streamlining LLM Application Evaluation2 days ago

LangChain Blog

1 day ago