Search tools, frameworks, libraries, categories and domains
Evaluation toolkit for RAG and agent pipelines with metrics such as faithfulness and context recall.
pip install ragas
Package ragas, last 26 weeks.
ragas
Other tools in the same categories.
langfuse/langfuse
Open-source LLM engineering platform for traces, evals, prompt management and metrics; now part of ClickHouse.
langchain.com
LangChain's hosted platform for tracing, evaluating and monitoring LLM apps and agents, framework-agnostic.
Arize-ai/phoenix
Open-source AI observability and evaluation tool built on OpenTelemetry and OpenInference.
braintrust.dev
Evaluation-first AI platform where production traces become datasets and regression tests.
promptfoo/promptfoo
CLI and library for testing prompts, comparing models and red-teaming LLM apps in CI; acquired by OpenAI.
confident-ai/deepeval
Pytest-style framework for unit-testing LLM outputs with research-backed metrics like G-Eval and faithfulness.
comet-ml/opik
Comet's open-source platform for tracing, evaluating and optimizing LLM apps and agents.
Helicone/helicone
Open-source LLM observability proxy: one-line integration for logging, costs, caching and rate limits.
traceloop/openllmetry
OpenTelemetry instrumentation for LLM providers, vector DBs and frameworks, by Traceloop.