Home AI data stack LLM observability and evals

LLM observability and evals

Traces, evals and monitoring for applications built on language models.

Showing 15 tools
Loading tools…

Confident AI (DeepEval)

Evals

LLM evaluation and observability platform built on open-source DeepEval: 50+ metrics for tests, CI and live production traces.

Visit site
Free; Pro Cloud from $19/month

Opik (Comet)

Observability

Open-source LLM observability and evaluation from Comet, with agent tracing, datasets, online scoring, prompt versioning, and optimization.

Visit site
From $20/user/mo

Lunary

Observability

LLM observability platform linking agent traces, chatbot conversations, feedback, prompt templates, and evaluation workflows.

Visit site
Free tier; Production from $49/month

Portkey

Gateway

AI gateway with provider routing, fallbacks, request logs, guardrails, and prompt management. Now branded Prisma AIRS AI Gateway.

Visit site
Free tier; Enterprise quote

Traceloop

Observability

LLM tracing, evaluations, and prompt management built around the Apache-2.0 OpenLLMetry SDK. The commercial platform has a free hosted tier and quote-based Enterprise deployment.

Visit site
Free open source; usage costs extra

TruLens

Evals

Open-source Python framework for evaluating RAG applications and AI agents with configurable metrics, OpenTelemetry traces, and version comparisons.

Visit site

Promptfoo

Evals

Open-source CLI for LLM evals and red teaming: YAML test suites, 60+ providers, CI gates and 157 attack plugins, now owned by OpenAI.

Visit site
Free open source; enterprise by email

Ragas

Evals

Open-source Python library for scoring RAG and LLM apps with LLM-judged metrics and generating synthetic test questions from your documents.

Visit site

Galileo

Evals

LLM and agent eval platform with 31 built-in metrics, Luna-2 judge models and production Signals, now sold as Splunk Agent Observability.

Visit site
Free; Pro from $60/mo

Weights & Biases Weave

Observability

LLM observability and evaluation toolkit for tracing AI applications, comparing outputs, and turning failures into repeatable tests.

Visit site

Helicone

Observability

Proxy-based LLM observability gateway with one-line setup; hosted product in maintenance mode since the March 2026 Mintlify acquisition.

Visit site

Arize AX

Observability

Enterprise LLM/ML observability (AX) with Phoenix, its ungated open-source tracing and evals core; deep drift and embeddings analysis.

Visit site
12