Disruption with GPT 5.3 Codex

What to do

• Check whether you use GitHub Copilot AI Model Providers. Check your stack

  Paste a package.json, lockfile, Terraform or a list of domains at down.guide/stack to see every provider you depend on.

• Confirm your retries, queues and scheduled jobs caught up after the 2h 33m it lasted.

  Anything that failed between 2026-07-20 16:03 UTC and 2026-07-20 18:37 UTC may need a replay.

• Subscribe to GitHub's feed to hear about the next one first. RSS feed

  https://down.guide/feed?provider=github works in Slack (/feed subscribe), Discord feed bots and any RSS reader.

Updates from GitHub 6

1. investigatingJul 20, 2026, 4:03 PM UTC

   We are investigating reports of degraded performance for Copilot AI Model Providers

   provider

2. investigatingJul 20, 2026, 4:04 PM UTC

   We are investigating GPT Codex 5.3 performance.

   provider

3. investigatingJul 20, 2026, 4:39 PM UTC

   We identified that an upstream provider is returning errors for GPT 5.3 Codex requests, causing some users to experience failed or interrupted responses. Traffic is being automatically rerouted to mitigate impact. We are working with the upstream provider to resolve the underlying issue. We recommend using a different model at this time while we resolve the problem.

   provider

4. investigatingJul 20, 2026, 5:21 PM UTC

   The upstream provider continues to return intermittent errors for GPT 5.3 Codex requests. Some users may experience failed or interrupted responses when using this model. Traffic continues to be automatically rerouted to reduce impact, and we are actively working with the upstream provider on a resolution. In the meantime, selecting an alternative model will avoid disruption.

   provider

5. investigatingJul 20, 2026, 6:25 PM UTC

   We are seeing signs of sustained recovery from the upstream provider for GPT 5.3 Codex requests. Requests are succeeding as expected. We are awaiting confirmation from the provider that the issue is fully mitigated before resolving this incident.

   provider

6. resolvedJul 20, 2026, 6:37 PM UTC

   Between 06:39 and 18:11 UTC on July 20, 2026, the Copilot service experienced a degradation of the GPT 5.3 model due to an issue with our upstream provider. The upstream model provider returned intermittent errors for GPT 5.3 Codex requests, which caused some responses to fail. Auto mode requests that had selected GPT 5.3 Codex were also impacted. On average about 2% of GPT 5.3 Codex requests failed during this window. Copilot automatically routed eligible traffic away from the impacted provider to reduce customer impact. No other models were impacted. We worked with the upstream provider throughout the incident and confirmed sustained recovery before resolving.

   provider

Corrections

None so far.

Something wrong on this page? Report an error or send us a note. Every fix is listed here.