Observability

OpenTelemetry-native, open-source tracing and evaluation for LLM and agent applications

Eval-first platform — datasets, scorers, CI gating, and logging for agent changes

Open-source LLM and agent observability — traces, evals, prompt management, and cost tracking

Tracing, evals, and prompt management with the deepest LangChain and LangGraph integration

Error and performance monitoring with AI agent tracing, plus a remote MCP server

Session replay, cost tracking, and failure analysis purpose-built for agent runs

One-line proxy for LLM logging, caching, rate limits, and cost analytics

OpenTelemetry-native agent observability with trace compression and low-volume pricing

Tracing, evaluation, and monitoring for LLM apps inside Weights and Biases