Model API status / OpenAI

OpenAI API Normal

The official API endpoint for GPT models. Both Responses and Chat Completions are served from api.openai.com.

Probed from Cloudflare edge DFW · updated 1 min ago (10/17 19:45)

Current response76 ms
24h median latency76 ms
24h P95172 ms
24h reachability100%
30-day reachability100%

Past 24 hours Hourly median latency. Red marks hours with failures

24 hours ago12 hours agoNow

Past 30 days One cell per day. Hover for details

30 days agoToday

Measured in mainland China

By carrier

APIAllChina UnicomMobile
OpenAI API11 samples300 ms成功 9%300 ms成功 11%Insufficient samples

Values are the median round-trip time over a reused connection. “Success” is the share of requests that received a response within 6 seconds.

Official incidents

OpenAI status page, last 30 days

  1. Critical
    Widespread feature degradation OpenAI · 10/16 00:29 · Lasted 2.5 hr
  2. Minor
    Delayed Costs data in the Compliance API OpenAI · 10/15 01:00 · Lasted 42 hr
  3. Minor
  4. Minor
    Degraded performance in Codex and ChatGPT Work OpenAI · 10/14 15:05 · Lasted 2.8 hr
  5. Minor
    Degraded performance in GPT-6.1 Sol Codex and Work mode OpenAI · 10/13 16:10 · Lasted 8.4 hr
  6. Critical
    Workspace admins may be unable to manage plugins OpenAI · 10/10 05:48 · Lasted 1.6 hr
  7. Critical
    Some ChatGPT content may be unavailable on Android OpenAI · 10/10 05:00 · Lasted 3.4 hr
  8. Minor
    Delayed Costs data in the Compliance API OpenAI · 10/10 01:23 · Lasted 3 days
  9. Minor
  10. Minor
    Elevated errors with GPT-5.6 Instant in FedRAMP workspaces OpenAI · 10/08 07:42 · Lasted 21 min
  11. Minor
  12. Minor
    Dot in Codex may fail to create new threads OpenAI · 10/08 00:31 · Lasted 1.6 hr
  13. Critical
  14. Minor
    Workspace Agents Degraded and Responses Impacted OpenAI · 10/07 09:12 · Lasted 1.8 hr
  15. Minor
    Computer Use Is Experiencing Issues OpenAI · 10/07 01:15 · Lasted 2.8 hr
  16. Minor
    Errors processing large PDFs in the Responses API OpenAI · 10/06 23:45 · Lasted 2.1 hr
  17. Minor
    Elevated error rates in ChatGPT Go conversations OpenAI · 10/06 21:35 · Lasted 1.1 hr
  18. Minor
  19. Minor
  20. Info
    Elevated error rates in Codex Cloud OpenAI · 10/06 01:58 · Lasted 6 hr
  21. Minor
    Elevated errors in ChatGPT conversations OpenAI · 10/06 00:58 · Lasted 7 hr
  22. Info
    Admin Console Intermittently Inaccessible OpenAI · 10/05 23:50 · Lasted 5.3 hr
  23. Minor
    Issues saving and editing Pages OpenAI · 10/05 20:44 · Lasted 20 min
  24. Minor
    Elevated Work Mode errors OpenAI · 10/05 15:03 · Lasted 19 hr
  25. Minor
    Issues with data analysis and file creation in ChatGPT OpenAI · 10/03 00:03 · Lasted 48 min

FAQ

About OpenAI API

Is the OpenAI API working right now?

As of 10/17 19:45, api.openai.com is reachable from Cloudflare edge nodes with a response time of 76 ms. Over the past 24 hours, reachability was 100% and median latency was 76 ms. The official status page currently shows: All Systems Operational.

What is the latency to the OpenAI API from mainland China?

过去 7 天中国大陆访客共测得 11 个样本，请求成功率 9%，连接复用后往返中位数 300 ms，首次连接中位数 765 ms。

How does cdnping monitor the OpenAI API?

Every 5 minutes, we send a request without an API key from Cloudflare edge nodes to https://api.openai.com/v1/models. Any status code below 5xx, such as 401, counts as reachable. We record the time from sending the request to receiving the response headers. See the methodology page for details.

Other APIs

cdnpingAGI PlanClaude APIGemini APIxAI APIMistral APIGroq APIOpenRouterDeepSeek APIZhipu BigModelKimi Open PlatformAlibaba Cloud Model StudioVolcengine ArkSiliconFlow

Related articles

Further reading