Live / Groq / Incident
Groq: llama-3.3-70b-versatile Performance Issue
Groq · groq.comResolved2025-09-08 23:01 UTC
Public — what happened
Summary
Groq reported on September 8, 2025 at 21:08 UTC that some requests to its llama-3.3-70b-versatile model were producing lower-quality responses. The company said it had found the problem and was working on a fix, noting that users could keep seeing the issue until the fix was fully rolled out. By 22:56 UTC Groq said the fix was in place and the poor output had stopped, and it marked the incident resolved at 23:01 UTC. Groq did not describe a cause.
Timeline
- Started — 2025-09-08 21:08 UTC
- Identified — 2025-09-08 21:08 UTC
- Monitoring — 2025-09-08 22:56 UTC
- Resolved — 2025-09-08 23:01 UTC · 1h 53m after it started
With SquidGrid — where it starts on the route
Independent path health
The status page is the symptom. SquidGrid measures the paths that reach Groq against their own baselines — whether the routes carrying your traffic degraded, at which handoff, and who owns that stretch.
🔒 Sign in to seeCustomers get the live Provider Assessment Report for Groq: its full incident track record, what tends to break, and recent public reports of problems.
Included with any paid plan.
Open the Groq report →
More from Groq