GH Repository · Arize-ai
phoenix
AI Observability & Evaluation
- stars
- 11,464
- 30-day movement
- starts with the next reading
- Related entries
- 61
- Connections
- 5
python/uvpythonPythonsmolagentsllamaindexlangchainllmopsllm-evaluationmakedockerai-monitoringai-observabilityllmsllm-evalanthropicdatasetsagent-frameworkaiengineeringprompt-engineeringevalsagentsopenai
Phoenix is an AI observability and evaluation tool from Arize, written in Python. Its topics indicate it monitors LLMs and agents, runs evaluations, and integrates with frameworks such as OpenAI, Anthropic, LangChain, LlamaIndex, and smolagents.
Reach for it when you need to trace, monitor, and evaluate LLM and agent behavior in one tool.
Use it to
- Trace and monitor LLM and agent runs
- Run LLM evaluations on datasets
- Debug agent behavior across frameworks
- Support prompt engineering workflows
For AI engineers doing LLMops, monitoring, and evaluation
- Role
- agent-framework
- Language
- Python
- Licence
- custom (licence file present)
- Forks
- 1,130
- Open issues
- 857
- Last push
- 2026-09-15
- Latest release
- v0.0.2rc0 · 2023-02-17
topicsllmopsobservabilityevaluationmonitoringagentspython