Arize Phoenix
Open-source and managed observability for AI agents, covering tracing, evals, datasets and experiments.
How the 7.2 was reached
- Quality of output7/10
11,654 GitHub stars, active pushes, 1,170 forks; vendor claims of 5M monthly downloads; 1,039 open issues and unclear licence temper it. github.com
- Ease of use7/10
Free AX plan needs no credit card; open-source option available; only thin evidence on docs or onboarding. arize.com
- Pricing value8/10
Free tier with 25k spans/month, Pro at $50/month, startup pricing; open-source self-host option exists. arize.com
- API and integration quality8/10
OpenTelemetry instrumentation, many agent framework integrations, and a remote MCP server for Claude Code and Cursor. github.com
- Solves the problem it claims to solve6/10
Tracing, evals, datasets and experiments are listed and adoption is solid, but no independent proof of delivery beyond GitHub adoption. github.com
7.2/10 is the mean of the 5 criteria that apply. Scored on the five-criteria rubric v1, 29 September 2026. How scoring works
Pricing
- 10 issues/month
- 25k trace spans/month
- 1 GB ingestion
- 15 days retention
- 25 issues/month
- 50k trace spans/month
- 10 GB ingestion
- 30 days retention
Key Features
- Tracing and evaluation of agent runs
- Datasets, experiments, playground and prompt management
- OpenTelemetry-based instrumentation
- Integrations with OpenAI Agents SDK, Claude Agent SDK, LangGraph, CrewAI, LlamaIndex and DSPy
- Remote MCP server for Claude Code and Cursor
Pros & Cons
Pros
- Free tier needs no credit card, and an open-source self-host option exists
- 11,654 GitHub stars and 1,170 forks show real adoption
- OpenTelemetry base and many framework integrations
- MCP server lets coding tools query traces and experiments
Cons
- 1,039 open issues on GitHub
- Licence is unclear
- Little evidence on docs and onboarding quality
- No independent proof of results beyond adoption; volume claims are the vendor's
Arize Phoenix suits teams building agents who want tracing and evals with a free or self-hosted start. Check the licence before adopting the open-source code. Teams that need hard evidence of onboarding quality may want to trial it first.
Try Arize Phoenix →Added to scored.tools on
Competitors to Arize Phoenix
Other tools in the coding category worth comparing.
LangGraph
8.0/10Low-level orchestration framework for building, managing, and deploying stateful AI agents.
LangChain
8.4/10Open source framework for building LLM-powered agents and applications with prebuilt architecture and integrations.
Evidently AI
7.6/10Open-source ML monitoring and testing framework for data drift and model performance
Haystack
8.4/10Open-source framework for building production-ready AI agents, RAG systems, and context-engineered applications.