Model API status / Groq
Groq API Normal
Groq, known for LPU inference, offers an OpenAI-compatible API.
Probed from Cloudflare edge DFW · updated 2 min ago (10/17 19:45)
Past 24 hours Hourly median latency. Red marks hours with failures
Past 30 days One cell per day. Hover for details
Measured in mainland China
By carrier
| API | All | China Unicom | Mobile |
|---|---|---|---|
| Groq API11 samples | 292 ms成功 100% | 301 ms成功 100% | 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
Groq status page, last 30 days
-
Critical
Model Performance or Availability Issue Groq · 10/13 02:17 · lasted 18 min
-
Minor
GroqCloud Console Errors Affecting Multiple Regions Groq · 10/12 11:37 · lasted 23 min
-
Minor
Model Performance or Availability Issue - minimaxai/minimax-m2.7 Groq · 10/10 18:09 · lasted 6.1 hr
FAQ
About Groq API
Is the Groq API working right now?
As of 10/17 19:45, api.groq.com is reachable from Cloudflare edge nodes with a response time of 106 ms. Over the past 24 hours, reachability was 100% and median latency was 90 ms. The official status page currently shows: All Systems Operational.
What is the latency to the Groq 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 100%，连接复用后往返中位数 292 ms，首次连接中位数 926 ms。
How does cdnping monitor the Groq API?
Every 5 minutes, we send a request without an API key from Cloudflare edge nodes to https://api.groq.com/openai/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.