T
Outage recap · Tavus

Tavus outage — September 14, 2026

On September 14, 2026, Tavus had a reported incident on Tavus's official status page. Below is what was reported — impact, timing, and the AI services the same upstream outage could affect.

Conversations using Gemma-4 are experiencing delays

minor
Started
Sep 14, 1:58 PM EDT
Resolved
Sep 14, 3:02 PM EDT
Duration
1h 5m

Tavus's updates (8)

  1. investigating · Sep 14, 1:58 PM EDT

    We are aware of an issue currently impacting tavus conversations, we are investigating.

  2. investigating · Sep 14, 2:03 PM EDT

    It appears this issue is coming from one of our downstream providers, Crusoe. We're continuing to investigate full scope of conversations affected.

  3. investigating · Sep 14, 2:14 PM EDT

    Starting at 17:40 UTC, conversations using our default LLM began experiencing long delays and unresponsive replicas. A GPU provider is returning errors and our automatic failover is recovering most turns, but with significant added latency. We are routing traffic to a backup provider.

  4. identified · Sep 14, 2:18 PM EDT

    This is affecting gemma-4; we recommend our customers switch to a different LLM for immediate needs.

  5. identified · Sep 14, 2:25 PM EDT

    The issue is isolated to our default LLM (gemma-4), served by a GPU provider that began returning errors at 17:40 UTC. As a temporary mitigation we are routing gemma-4 traffic to an alternate model. Customers on gemma-4 may notice a change in response style while this is in effect; we will revert once the provider recovers.

  6. identified · Sep 14, 2:42 PM EDT

    We've confirmed the issue originates with our GPU provider, who has declared a networking incident on their side. We're testing mitigations to route gemma-4 traffic elsewhere. Conversations using other LLMs are unaffected. If you need stability now, you can point your persona at a different LLM.

  7. monitoring · Sep 14, 2:55 PM EDT

    Our upstream GPU provider has applied a mitigation and gemma-4 error rates have returned to normal. We have alternate routing paths staged and ready to enable if the issue recurs. We are monitoring before marking this resolved.

  8. resolved · Sep 14, 3:02 PM EDT

    This incident is now resolved.

What our checks recorded

We checked Tavus 13 times from Tavus's start time to its resolution, once every five minutes. Every one of them reached Tavus's site, so this incident did not show up as a failed page load. Our check loads the public site; it cannot see an API error or a broken feature. Median response time was 102 ms, against 133 ms over the 24 hours before. Tavus posted the incident at Sep 14, 1:58 PM EDT.

Updates are Tavus's own words from its status page (status: resolved); we never add an inferred cause. The vendor's incident page ↗

AIDownCheck is not affiliated with Tavus. This recap summarizes incidents published on Tavus's official status page.