Skip to main content

Autoevals

Pre-built scoring functions for evaluating LLM outputs, including factuality, similarity, and answer relevance scorers. Use it alongside the main SDK to score evaluations without writing your own scorers from scratch. API reference · npm: autoevals

LangChain

Braintrust’s LangChain.js integration. Adds Braintrust tracing to LangChain workflows so chains, agents, and runnable steps appear as spans in your project. API reference · npm: @braintrust/langchain-js

OpenAI Agents

Braintrust’s integration for the OpenAI Agents.js framework. Traces agent runs, tool calls, and handoffs, with no manual instrumentation required. API reference · npm: @braintrust/openai-agents

OpenTelemetry

Braintrust’s OpenTelemetry JS integration. Exports OTel-compatible traces from your application to Braintrust, useful when your app is already instrumented with OpenTelemetry. API reference · npm: @braintrust/otel