Arize Phoenix
Arize's source-available AI observability platform for tracing, evaluation, datasets and experiments.
Sources and review
Checked against public documentation on . Suitability guidance is our editorial judgment; this is not a performance benchmark or an endorsement.
Documentation reviewed
Can I use this for my work?
Phoenix supports tracing, evaluations, datasets, experiments and prompt work. Its current server license is Elastic License 2.0; Arize AX is a separate managed offering.
- Availability
- Available. Released Phoenix packages; individual agent or integration features can have separate beta stages. This record does not cover Arize AX plans.
- How to use it
- Web app; API; Library / CLI. Local or self-hosted deployment with supported runtime/database; inspect current deployment documentation.
- Processing and data
- On-device option; Self-hosted option. Trace storage belongs to the selected Phoenix deployment. LLM evaluators can send content to a separate configured model provider; configure retention and input redaction.
- Costs and limits
- Self-host software; infrastructure and evaluator/model costs
- Outputs
- Trace views, datasets, evaluator scores and comparative experiments; these support investigation rather than certifying an AI application.
ELv2 is source available, not an OSI open-source license. Hosting and evaluator/model calls can cost money. Review trace contents and retention; no runtime or security effectiveness was tested.
Compare these facts with another tool →
Publisher documentation was reviewed. ASE has not independently tested this workflow, its outputs, billing, privacy controls or accessibility.
Teams that want self-hosted tracing and evals built on OpenTelemetry standards.
You need an OSI open-source licence; Phoenix uses the Elastic License 2.0, which restricts offering it as a managed service.
About Arize Phoenix
Arize Phoenix is an AI observability platform for experimentation, evaluation and troubleshooting. It traces LLM applications with OpenTelemetry-based instrumentation, runs LLM-based response and retrieval evaluations, manages versioned datasets and experiments, provides a prompt playground and prompt management, and includes an AI agent for debugging traces. It runs locally or self-hosted; Arize AX is the company's managed product.
Arize Phoenix is listed in this index written primarily in Python, with 12k stars on GitHub. Its latest recorded repository push was in November 2026; that is not an ASE editorial review date.
pip install arize-phoenix