Demo workspace ready Local + provider comparison cockpit

Home

Pick a prompt, compare models, and turn the best result into a repeatable evaluation.

stadia_controllerOpen Arena
Ready models
6
memory
Active skills
6
hub
Demo evaluations
5
fact_check
Latest winner
Qwen3 14B
emoji_events

Latest evaluations

View all
API support classifierDeveloper/API - GPT-5 mini won with 95% JSON Medical triage safetyMedical - Claude Sonnet won with 93% Safety Executive KPI insightBusiness - Claude Sonnet won with 92% Business

Start here

Compare a prompt in Arenaarrow_right_alt Create an evaluationarrow_right_alt Review modelsarrow_right_alt Manage skillsarrow_right_alt