Skip to content
← Tools / Skills
ToolArize AI/10 trustVerified
Phoenix logo

Phoenix

AI observability for traces, evals, and experiments — Elastic-licensed.

Visit official site →

About

Phoenix is Arize's AI observability platform for experimentation, evaluation, and troubleshooting: OpenTelemetry-based tracing of LLM app runs, LLM-as-judge evals, versioned datasets and experiments, a prompt playground for comparing models, and PXI — a built-in AI agent for debugging traces. It installs via pip/conda/Docker and is MCP-enabled. License note: the repo is under Elastic License 2.0 — source-available, not OSI open source, despite 'open-source' marketing language. Managed production workflows upsell to Arize AX. 11.8k stars.

What it can do

  • **Tracing**: OpenTelemetry-based LLM run tracing with span inspection
  • **Evaluations**: response and retrieval evals with LLM judges
  • **Datasets & experiments**: versioned examples; track prompt/model/retrieval changes
  • **Playground**: optimize prompts, compare models, replay traced calls
  • **PXI agent**: built-in AI engineer for debugging traces and iterating prompts
  • **MCP + Cursor**: MCP server and one-click Cursor integration

Setup

```bash

pip install arize-phoenix

```

```python

import phoenix as px

px.launch_app() # local UI for traces

```

Instrument with OpenTelemetry (auto-instrumentation for OpenAI/LangChain/LlamaIndex). Docker + Helm for self-host. Docs: arize.com/docs/phoenix.

Prerequisites

Python 3.9+. Optional: Docker for production self-host; API keys for the models you trace.

Use cases

observabilityevalstracingprompt engineering

Works with

customlangchainGeneric API

Trust score — 8/10

/10 trustVerified

Scored 8/10 (verified tier, metrics-v1). Evidence: 11,808 GitHub stars (log-scaled vs catalog peers); last push 0d ago; 100 releases/12mo; 1158 open issues. Manual review: Arize observability (OTel tracing, evals, datasets, playground, PXI agent); 11.8k stars; managed Arize AX is the monetization funnel. License: Elastic License 2.0 — source-available, NOT OSI open source (despite 'open-source' marketing).

★ 11,808 GitHub stars

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
7.7/10
Maintenance · recency, releases, issue load
8.2/10
Openness · license + docs
7.2/10
Manual review · sanity read + red-flag check
8.0/10
  • · 11,808 GitHub stars (log-scaled vs catalog peers)
  • · last push 0d ago; 100 releases/12mo; 1158 open issues
  • · manual review: Arize observability (OTel tracing, evals, datasets, playground, PXI agent); 11.8k stars; managed Arize AX is the monetization funnel

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.