AI API status

Whether each provider’s API is up right now, from their official status pages, checked every five minutes.

All monitored APIs are operational.

  • Anthropic Operational

    Last 30 days · 98.24% of checks operational Checked Oct 27, 2026 07:05

    Official status page
  • OpenAI Operational

    Last 30 days · 94.66% of checks operational Checked Oct 27, 2026 07:05

    • Elevated error rates for most hosted Code Reviews and Security Reviews on private GitHub repositories · monitoring
    Official status page
  • Google Operational

    Last 30 days · 100% of checks operational Checked Oct 27, 2026 07:05

    Official status page
  • DeepSeek Operational

    Last 30 days · 99.77% of checks operational Checked Oct 27, 2026 07:05

    Official status page
  • Cohere Operational

    Last 30 days · 100% of checks operational Checked Oct 27, 2026 07:05

    Official status page
  • ElevenLabs Operational

    Last 30 days · 100% of checks operational Checked Oct 27, 2026 07:05

    Official status page
  • Runway Operational

    Last 30 days · 100% of checks operational Checked Oct 27, 2026 07:05

    Official status page
  • xAI Not monitored

    Its status page blocks automated reading (bot protection), so we cannot check it for you. Open it directly.

    Official status page
  • Mistral AI Not monitored

    Its status page blocks automated reading (bot protection), so we cannot check it for you. Open it directly.

    Official status page
  • Meta Not monitored

    This provider does not publish a public status page.

  • Black Forest Labs Not monitored

    This provider does not publish a public status page.

  • Moonshot AI Not monitored

    This provider does not publish a public status page.

  • Alibaba Cloud (Qwen) Not monitored

    This provider does not publish a public status page.

We relay what each provider publishes; we do not run our own probes. A provider can be slow or failing before its status page says so.

Frequently asked questions

Is the OpenAI, Claude or Gemini API down right now?

The table above shows the latest state each provider publishes, refreshed every five minutes. For the full incident report, open the provider’s official status page from its row.

What should I do when a provider is down?

Retry with exponential backoff, and for production systems keep a fallback model from another provider configured. The alternatives listed on each model page are a starting point.