Phoenix
Phoenix Arize is an open-source AI observability platform for LLMs.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for Phoenix, derived from its capabilities.
AIVSS 7.1 · High
View MAESTRO 7-layer threat model →Overview
Phoenix Arize is an open-source AI observability platform designed to trace, evaluate, and optimize language model (LLM) applications. Built on OpenTelemetry, it is vendor, framework, and language agnostic, providing flexibility and comprehensive visibility into the inner workings of AI applications. It supports pre-deployment evaluation and troubleshooting directly from a local machine, making it ideal for early-stage developers.
Key features
- Performance tracing, dashboards, LLM evaluation library, span-level visibility, prompt playground.
Use cases
- Troubleshooting LLM applications, optimizing prompt templates, reducing hallucinations, evaluating RAG relevance, monitoring AI performance.
Listing aggregated from aiagentsdirectory.com