Monitoring - We have implemented a fix for a configuration error which caused this issue. That fix is live so now the billing and usage dashboard data should be displaying normally for all who should have access. We're monitoring a backfill of all related billing events to ensure the usage display since this issue started reflect correctly in customer dashboards.
Oct 16, 2024 - 10:39 PDT
Update - Users should now see data reflected in their dashboards for any new usage or billing events. We are still working on resolving the missing historical usage and billing events.
Oct 14, 2024 - 14:14 PDT
Identified - We are currently investigating an issue where multiple users have reported no activity appearing in their billing and usage dashboards for certain models since October 10, 2024. Affected API customers are still able to use our models without disruption. Starting from ~1 PM PT today, users should start seeing data reflected again.
Oct 14, 2024 - 13:12 PDT
API ? Operational
90 days ago
99.82 % uptime
Today
ChatGPT ? Operational
90 days ago
99.82 % uptime
Today
Labs ? Operational
90 days ago
99.97 % uptime
Today
Playground ? Operational
90 days ago
99.94 % uptime
Today
Operational
Degraded Performance
Partial Outage
Major Outage
Maintenance
Major outage
Partial outage
No downtime recorded on this day.
No data exists for this day.
had a major outage.
had a partial outage.
Past Incidents
Oct 17, 2024
Resolved - We've implemented a fix and confirmed the elevated latency on GPT-4o mini is resolved. This started Saturday Oct 12 11pm PT and resolved on Thursday Oct 17 3am PT.
Oct 17, 14:40 PDT
Monitoring - We believe the issue has been mitigated. We're continuing to monitor the performance of this model.
Oct 17, 09:28 PDT
Investigating - We are currently investigating increased latency on our GPT-4o mini model in the API. ChatGPT does not appear to be impacted at this time.
Oct 16, 14:24 PDT
Resolved - This incident is now resolved across all impacted services. This incident started Thursday October 17th, 2024 at 10:21pm and was mitigated at 10:45pm Pacific time.
Oct 17, 10:51 PDT
Monitoring - A fix has been implemented and we are monitoring the results. The fix is applied to all impacted services across ChatGPT, custom GPTs, the API, and Assistants.
Oct 17, 10:43 PDT
Update - Users are currently unable to upload files in ChatGPT / custom GPTs. API and Assistants are also impacted at this time.
Oct 17, 10:33 PDT
Investigating - Users are currently unable to upload files in ChatGPT / custom GPTs. We are currently investigating this issue. API/Assistants are not impacted at this time.
Oct 17, 10:31 PDT
Oct 16, 2024
Oct 15, 2024
Resolved - On Tue, Oct 15, we experienced degradation in the Realtime API from 11:45 PM to 12:45 AM GMT (Oct 16).
Oct 15, 16:30 PDT
Oct 14, 2024
Oct 13, 2024

No incidents reported.

Oct 12, 2024

No incidents reported.

Oct 11, 2024

No incidents reported.

Oct 10, 2024

No incidents reported.

Oct 9, 2024

No incidents reported.

Oct 8, 2024
Resolved - This incident has been resolved.
Oct 8, 11:01 PDT
Monitoring - A fix has been implemented and we are monitoring the results.
Oct 7, 22:02 PDT
Identified - The issue has been identified and a fix is being implemented.
Oct 7, 14:22 PDT
Investigating - Recently created fine-tuned models are not showing up in the "Model Selection" dropdown when choosing a base model to create a fine-tuning job from. This issue is only in the UI, customers can create fine-tuning jobs based off of existing fine-tunes via the API. This issue appears to only impact fine-tuned models created in the default project.
Oct 7, 13:39 PDT
Oct 7, 2024
Oct 6, 2024

No incidents reported.

Oct 5, 2024

No incidents reported.

Oct 4, 2024
Resolved - This incident has been resolved.
Oct 4, 06:22 PDT
Monitoring - A fix has been implemented and we are monitoring the results.
Oct 4, 06:12 PDT
Update - We are continuing to investigate this issue.
Oct 4, 06:03 PDT
Investigating - We're experiencing an elevated level of API errors for calls with vector stores
Oct 4, 05:59 PDT
Oct 3, 2024

No incidents reported.