Live status · 14 services · checked every 5 minutes
Is it down, or is it you?
Live checks on the registries, AI APIs and hosts developers depend on, plus what the official status pages say. Open a service for its history and what to do while it is down.
Right nowAll 14 services are answering. Every check passed on the last round, and no official status page reports an open problem.
Uptime, side by side
Uptime is the share of our 5-minute rounds in which every critical endpoint answered as expected. We started checking on Oct 24, 2026, so longer windows fill in over time.
| Service | Now | Latency | 24 hours | 7 days | 30 days | Status page |
|---|---|---|---|---|---|---|
| crates.io | ●Up | 38 ms | 100% | 100% | 100% | official |
| docs.rs | ●Up | 144 ms | 100% | 100% | 100% | official |
| GitHub API | ●Up | 218 ms | 100% | 100% | 100% | official |
| npm | ●Up | 231 ms | 100% | 100% | 100% | official |
| PyPI | ●Up | 158 ms | 100% | 100% | 100% | official |
| OpenAI API | ●Up | 237 ms | 100% | 100% | 100% | official |
| Anthropic API | ●Up | 186 ms | 100% | 100% | 100% | official |
| Gemini API | ●Up | 168 ms | 100% | 100% | 100% | official |
| OpenRouter | ●Up | 207 ms | 100% | 100% | 100% | official |
| Vercel | ●Up | 314 ms | 100% | 100% | 100% | official |
| Cloudflare | ●Up | 190 ms | 100% | 100% | 100% | official |
| App Store Connect | ●Up | 372 ms | 100% | 100% | 100% | official |
| Steam store | ●Up | 316 ms | 100% | 100% | 100% | official |
| Hugging Face | ●Up | 219 ms | 100% | 100% | 100% | official |
Recent incidents
From the official status pages, newest first. Each service page also lists the outages our own checks saw.
- Anthropic API · 37 minElevated errors for Claude Fable 5 and Claude Fable 5.1
- Cloudflare · 2.1 hNetwork Performance Issues in Ashburn
- Anthropic API · 6 minElevated errors for Claude Opus 5.5
- Anthropic API · 34 minElevated errors for Claude Fable 5.1, Sonnet 5.5, and Haiku 5.5
- OpenAI API · 33 minHelp Center Chat Support Degraded
- Cloudflare · 2.9 hNetwork connectivity issues affecting services in North America
- Cloudflare · 5.2 hDelayed, missed, and duplicated Workers Cron Triggers
- GitHub API · 3.9 hDisruption with Dependabot Alerts
- Cloudflare · 21 minCloudflare Access Availability Issues
- GitHub API · 36 minDelayed Actions status check results
- OpenAI API · 2.3 hDelays in customer support responses
- OpenAI API · 4.5 hNew Ad Accounts failing to provision admins
How we check
Every 5 minutes a server in Washington, D.C. (Vercel, iad1) sends each service a small public request: the endpoints your tools actually hit, like the crates.io sparse index, the npm registry and the PyPI simple index. For keyed APIs (OpenAI, Anthropic, Gemini, App Store Connect) we send no key, so a fast 401 or 403 means the front door is answering. We never use an API key. A failed check is retried once after 2 seconds before it counts. We also read each provider's official status page and show the worse of the two.
What this can't see: problems in one region far from Washington, or errors that only happen with a valid key and a real request. For those, the official page is the authority, and we link it on every service page.
Use it from code
The data is free and open: JSON for every service, one service with ?id=crates-io, an RSS feed of incidents, and a live badge for your README on each service page.
Up next
The most checked service: What cargo still does when crates.io is down, and the three-line mirror config that gets you building again.Is crates.io down?Live crates.io status, uptime and what to do while it is down.Read it next →