Is Groq down?
Groq
No — Groq is working normally
Our sources see nothing wrong. If it still fails for you, the cause is most likely on your side — the tips further down cover the usual suspects.
What our sources say
- Official status page
- Working
- Our test requests, last hour
- Not running for this service
- Traffic through OpenRouter
- 0 of 3 models failing, 100% of requests succeeded in 30 min
- Reports from visitors, last hour
- 0
Parts of the service
- API Working
- llama-3.1-8b-instant Working
- llama-3.3-70b-versatile Working
- openai/gpt-oss-120b Working
- openai/gpt-oss-20b Working
- whisper-large-v3 Working
- whisper-large-v3-turbo Working
- Canopy Labs Orpheus Arabic Saudi Working
- Canopy Labs Orpheus English Working
- meta-llama/llama-prompt-guard-2-22m Working
- meta-llama/llama-prompt-guard-2-86m Working
- openai/gpt-oss-safeguard-20b Working
- minimaxai/minimax-m2.7 Working
- qwen/qwen3.8-27b Working
- Website Working
Last 24 hours
24 h agonow
Last 30 days
30 days agoyesterday
27 of 30 days with data had no problems.
Recent incidents
- Partly down
Model Performance or Availability Issue
Resolved after 18 mins
- Slow or glitchy
GroqCloud Console Errors Affecting Multiple Regions
Resolved after 23 mins
- Slow or glitchy
Model Performance or Availability Issue - minimaxai/minimax-m2.7
Resolved after 6 hrs 5 mins
About this service
Groq serves open models on its own LPU chips. The free tier has tight per-minute limits, so a burst of 429 errors is far more often a limit than an outage.
Other model APIs right now
Groq: common questions
Is Groq down, or just one model?
Groq serves many models, and they rarely fail all together. If one model returns errors, try another of a similar size: if that works, the trouble is limited to that model or the hardware it runs on. When unrelated models fail at the same moment, the platform itself is struggling.
What can I do while Groq has problems?
Most model hosts speak the OpenAI-compatible API, so moving traffic to another host that serves the same open model is often just a different base URL and key. Set that fallback up before you need it, and meanwhile retry 5xx errors with growing pauses.
Why am I getting 429 errors from Groq?
A 429 means you've reached a request, token or concurrency limit of your plan, not that the service is down. Limits differ from model to model: spread requests out, upgrade the plan or pick a model with more headroom.
Why does it work for a colleague and not for me?
Requests can land on different hardware depending on the account and the model, so two people can see different results in the same minute. Your key's limits, the exact model ID and the region you call from all play a part.
How do you know whether Groq is down?
For Groq we combine Groq's official status page, read every minute, the share of successful requests to Groq's models on OpenRouter and reports from people who press the button on this page. The most serious signal that is still fresh sets the verdict, and stale sources are ignored rather than trusted. The page refreshes by itself; the exact thresholds are on our “How we check” page.