20 open-source AI agent eval tools worth knowing in 2026
MaryamMiradi · x · 2026-11-04
A curated map of 20 open-source tools for evaluating production AI agents, grouped into five categories:
- Eval frameworks: DeepEval, Inspect AI, AutoEvals, Evidently
- RAG + response evals: Ragas, TruLens, LlamaIndex Evals, LangChain Evals
- Tracing + observability: Langfuse, Arize Phoenix, Opik, Helicone
- Testing + red teaming: Promptfoo, Giskard, garak, PyRIT
- Agent benchmarks: AgentBench, τ-bench, BrowserGym, SWE-bench
The author's core point: production agents must be evaluated far beyond the final answer—tool calls, retrieval quality, reasoning paths, failed steps, latency, cost, safety, and regressions. The best-performing agent isn't necessarily the most reliable one. A follow-up head-to-head benchmark of the eval frameworks is teased.
More from coding & agent
- Sentry's AI Agent goes head-to-head with Semgrep bot in a real PR review — zeeg · 2026-11-04
- Snowflake open-sources Arctic Platform, a modular post-training framework for LLMs — StasBekman · 2026-11-04
- Sentry Engineer Shares Bots Arguing on a PR: Security Bot Flagged, AI Bot Refused and Resolved — zeeg · 2026-11-04
- Zed's zeeg: Sentry tools now globally disclosed and already shipped — zeeg · 2026-11-04
- Particle-1 lands in OpenClaw with ~45% fewer output tokens than DeepSeek V4.1 Flash — garrytan · 2026-11-04
- Grok bots vs OpenClaw: multi-LLM agents race to forecast US elections live — alvelda · 2026-11-04