AI observability and evaluation platform to trace, evaluate and improve LLM agents in production, with an open-source Phoenix core.
What it does
Arize is an AI engineering platform for observability and evaluation of LLM apps and agents. Its managed Arize AX product and open-source Phoenix tool let teams trace requests, run evals at scale, debug failures and continuously improve prompts and agents. It is built on OpenInference and OpenTelemetry.
How to use: Integrate Arize AX with your AI development and production pipelines using OpenTelemetry for seamless visibility. Use the platform to trace prompts, variables, tool calls, and agents, debug faster, and automate AI evaluation at every stage.
Core features
Agent and LLM tracing
Large-scale evaluations
Open-source Phoenix observability
Alyx AI engineering agent
OpenTelemetry-based instrumentation
Experiments and prompt playgrounds
Best for
→Debugging AI agents in production
→Measuring LLM output quality
→Catching regressions before deploy
Pricing
AX
Free
AX Pro
$50/mo
Toolspool rankingby monthly traffic