LLM Status Hub

Live API checks from multiple AWS regions, every minute

Loading…
How does this work?

We make a real API call to each model — the same call your app would make — from multiple AWS regions every minute. Latency is measured from request sent to full response received. A model is flagged degraded when its 1-hour average exceeds 1,000 ms or error rate rises above 1%, and down when it stops responding entirely.

Note: We measure API endpoints only. Web chat availability (ChatGPT.com, Claude.ai, etc.) is a separate product and is not tracked here.

Model Status Average over last hour
Latency: <1.0s ≥1.0s
Error Rate: <1.0% ≥1.0%
Down: No response
Model Latency (1h Avg) Errors (1h Avg) Updated
Loading…

Want alerts when models go down? Get Slack & webhook alerts →

Regional Latency Latest poll only
Latency: <500ms 500–999ms 1,000–1,999ms ≥2,000ms
Model US-East US-West
Loading…
Latency Trends Mean response time · 2-hour buckets

Know before your users do.
Prove it when they do.

🔒 Secured by Stripe 💸 No setup fees 🚀 Integration in minutes 💳 Manage billing ⏱️ Cancel anytime 🛡️ 14-day money-back guarantee
Running massive clusters? Contact us for Enterprise Volume Pricing.

Frequently Asked Questions

Integration guide is now available. View our Developer Documentation.