Iris Playground — Try Agent Eval in 60 Seconds

v0.4.0Iris v0.4 — LLM-as-Judge + citation verify + OTel + 9 MCP tools→

Product Playground Learn Compare Pricing Blog Open Source Roadmap

GitHub Get Started

Loading playground...

Get started with Iris

See what your AI agents are actually doing. One command. No SDK. Open-source core. MIT licensed.

Start building Star on GitHub

Iris

The agent eval standard for MCP. Open-source core. MIT Licensed.

Product

Features
Playground
Pricing
Blog
Quick Start
Roadmap

Learn

Agent Eval Guide
The Eval Tax
Eval Drift
The Eval Gap
Eval Coverage
EDD
The Eval Loop
Self-Calibrating Eval
Output Quality Score

Compare

Iris vs Langfuse
Iris vs LangSmith
Iris vs Helicone
Iris vs Braintrust
Iris vs Arize
Iris vs DeepEval
Iris vs Confident AI
Iris vs Patronus AI

Community

GitHub
Discord (coming soon)
npm
@iris_eval

© 2026 Iris. MIT Licensed.

Privacy Terms Security hello@iris-eval.com·Blog content is researched with AI assistance and reviewed for accuracy.