Not a ranking: each option is described by the situation it fits, with its trade-off, whether it's open source or free to start, whether AI coding agents can work with it, and how many makers' products use it for this job.
Open-source platform you can self-host, or a hosted eval product. Open-source tools keep traces on your own servers and cost nothing to run yourself, but you maintain them. Hosted eval platforms give you polished experiment and scoring workflows for a monthly bill, with your data in their cloud.
Promptfoo (Eval and test CLI): Running prompts and models against test cases from the command line or CI, including red-team tests for prompt injection and data leaks. Trade-off: A testing tool rather than production tracing, so you pair it with something that logs live traffic.
Best forTracing, prompt management and evals in one tool you can run on your own server for free.
Trade-offSelf-hosting means running Postgres, ClickHouse, Redis and object storage alongside it; otherwise it's a usage-based cloud plan.
Used by 33 makers' products for this
Best forGetting request logs, costs and latency by changing one base URL, with no SDK instrumentation.
Trade-offProxy logging sees individual requests well but agent steps and eval workflows less deeply; acquired by Mintlify in March 2026, so check its roadmap before building on it.
Used by 13 makers' products for this
Best forApps built on LangChain or LangGraph, where tracing works with almost no setup.
Trade-offPaid per seat beyond the free tier; self-hosting is enterprise-only.
Used by 11 makers' products for this
Best forEval-driven work — scoring outputs and comparing prompts and models side by side in experiments.
Trade-offClosed source, and the paid tier is priced for teams rather than hobby projects.
Used by 5 makers' products for this
Best forOpenTelemetry-based tracing and evals you can start locally in a notebook, then self-host or move to its cloud.
Trade-offSource-available under the Elastic License rather than a permissive open-source license.
No maker product tracked for this yet
Best forTracing, evaluations and prompt optimization in an Apache-licensed platform you self-host with Docker or Kubernetes, or use on Comet's cloud.
Trade-offSelf-hosting means running several services; smaller community than Langfuse.
No maker product tracked for this yet
Best forWriting LLM evals as pytest-style tests in Python, with ready-made metrics for hallucination, faithfulness and answer relevancy.
Trade-offThe metrics use an LLM as a judge, so every run costs tokens and scores vary slightly; dashboards and history are in the paid Confident AI platform.
No maker product tracked for this yet
Best forScoring RAG pipelines — how faithful answers are to the retrieved context and how relevant that context is — and generating test questions from your documents.
Trade-offA Python library, not a platform — no tracing of live traffic or UI, and LLM-judged metrics cost tokens per run.
No maker product tracked for this yet
Best forTracing, evaluating and versioning prompts for LLM apps in the same open-source platform many teams already use for ML experiments.
Trade-offGrew out of classic ML tooling, so the UI and concepts are broader than an LLM-only tool; you run the tracking server yourself unless you use a managed version.
No maker product tracked for this yet
