AI product Open source
Phoenix is an open-source AI observability platform developed by Arize for experimenting with, evaluating, and troubleshooting LLM applications and agents. It traces application runtime with OpenTelemetry-based instrumentation, supports response and retrieval evaluations, and provides versioned datasets, experiment tracking, prompt management, and a playground for comparing models, adjusting parameters, and replaying traced calls. Phoenix also includes PXI, an AI engineering agent for debugging traces and iterating on prompts, plus a remote MCP server for querying traces, datasets, and experiments from MCP clients. It can run locally, in containers, through Kubernetes and Helm, or in the cloud, and is distributed through the `arize-phoenix` Python package, Conda, Docker images, and related TypeScript and Python packages.
6 uses taken from transcripts — each links to the moment in the video.
Provides end-to-end tracing for agents.
Arize's open-source version; the skills and MCP are available in Phoenix as well, though online evals and dashboards are more polished in the enterprise edition.
Arize AI's open-source AI observability platform. It collects traces from AI apps built with any framework, stores evals alongside those traces, provides a UI to inspect spans, tool calls and LLM calls, and supports datasets, experiments and annotations. You run it locally on your laptop via uvx (localhost:6006) to keep data private, or on a shared server, and it works with OpenTelemetry/OpenInference instrumentation across frameworks like the Claude Agent SDK, Vercel AI SDK, Mastra, LangChain and OpenAI's agent SDK.
An eval library used to run tests against real data when choosing the right small model for a job. It was used to compare models like Llama 3.2B, Gemma and Qwen on summarization quality and latency, with a large model acting as judge.
Arize AI's open-source AI observability, experimentation and evaluation platform. It ingests traces via OpenTelemetry/OpenInference, maps Harbor datasets, experiments and rewards onto its own concepts, and provides versioned datasets, experiment comparison and trajectory visualization UIs.
An open-source observability platform developed by Arize AI that was used to build the company's agent harness and implement the Code Mode features.
6 in the library.