Model Radar

You'd reach for this when you need to find which free coding LLM is actually responding fastest right now. It pings over 134 models across 17 different providers, measures their latency, and ranks them so you can route your requests to whatever's most responsive at the moment. Useful if you're building something that needs to stay on free tiers but can't afford to wait around for slow API responses. The server exposes operations to run these latency checks and retrieve the ranked results, letting you programmatically pick the fastest available model instead of hardcoding a single provider that might be sluggish today.