Arize Phoenix
by Arize AI
What is Arize Phoenix? Arize Phoenix is a self-hostable observability and evaluation tool for LLM applications from Arize AI. It traces apps with OpenTelemetry-based instrumentation, runs LLM-graded evals on responses and retrieval, and manages versioned datasets and experiments. It is licensed under the Elastic License 2.0.
Editor note
A self-hostable tool for tracing and evaluating LLM apps: OpenTelemetry-based traces, LLM-graded evals for answers and retrieval, versioned datasets and experiments. It is released under the Elastic License 2.0; Arize sells a separate managed product (Arize AX).
- What it tests
- Your own LLM app or agent, on traces and test datasets.
- Original source
- github.com/Arize-ai/phoenix
- Last updated
- Repo last commit 2026-10-17
- License
- Elastic-2.0
When to use Arize Phoenix
- You want traces and eval scores in one self-hosted tool.
- Your stack uses common agent frameworks or providers that Phoenix's OpenInference instrumentation already supports.
Limitations
- The Elastic License 2.0 does not allow providing the software to others as a hosted or managed service; read the licence before building a product on it.
- It is a platform, not a benchmark: scores depend on the evaluators and datasets you set up.
How to run it
- pip install arize-phoenix, then phoenix serve (or run it with uvx or Docker).
- Instrument your app to send traces, then add evals, datasets and experiments.
pip install arize-phoenix
phoenix serveNotable results
- README (read 17 Oct 2026): Phoenix is self-hosted and licensed under the Elastic License 2.0; Arize offers a separate managed platform, Arize AX. [Official repository]
Sources
Reviews
No reviews yet.
Reviews from practitioners who sign in with GitHub open soon.
Related rubric libraries & frameworks evals
- Inspect Evals — Many benchmarks under one runner: coding, agents, safety, knowledge.
- OpenAI Evals — Building and running your own evals; community registry.
- LLMJury rubric library — LLM-as-judge rubrics for support, RAG, summaries, code and tone.
- Prompt Assay judge templates — Starter LLM-as-judge prompts.