Live  /  Groq  /  Incident

Groq: meta-llama/llama-4-scout-17b-16e-instruct & groq/compound Degraded Performance

Groq · groq.comMinor incidentResolved2025-12-02 20:02 UTC
Public — what happened

Summary

Groq said that on December 2, 2025, starting at 17:36 UTC, two of its AI models, meta-llama/llama-4-scout-17b-16e-instruct and groq/compound, were showing higher error rates and slower responses, while its other models kept working. The company applied a fix at 17:51 UTC, but by 18:06 UTC it reported that performance was still degraded and it was still looking for the cause. A second fix at 18:31 UTC improved things, and Groq declared the issue resolved at 20:02 UTC without publicly naming a cause.

Timeline

Source of record: provider status page. Summary written by SquidGrid from the provider’s published updates; times in UTC.

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 see
Customers 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

The full archive is on the Groq status & history page.

Related: Groq status & history. SquidGrid is not affiliated with Groq.