Relay Slow Performance

Response times have continued to improve and have remained stable within target over the past several weeks. This issue is now resolved.

Day: 6/10/26 — # of Requests: 3,880,501 — AVG Response Time (ms): 460.6
Day: 6/16/26 — # of Requests: 4,267,313 — AVG Response Time (ms): 326.8
Day: 6/23/26 — # of Requests: 3,512,703 — AVG Response Time (ms): 214.9
Day: 6/30/26 — # of Requests: 4,329,002 — AVG Response Time (ms): 188.0
Day: 7/07/26 — # of Requests: 3,675,799 — AVG Response Time (ms): 166.9

Overall, response time has improved 64% since we began tracking this issue. Pantheon continues to respond quickly and consistently across Relay and Junction. We will keep monitoring performance going forward. If you experience any specific issues, please submit a ticket with details (screenshots, timestamps, steps to reproduce, etc.) and we will investigate.

The table below shows one of the measured performance metrics, representing average request response time. Average response time is the measure of how long the Pantheon API requires to respond to an action. Lower response time translates into faster interaction. Overall, in the last two weeks, it has improved by 53%.

Day # of Requests AVG Response Time (ms)
6/10/26 3,880,501 460.6
6/16/26 4,267,313 326.8
6/23/26 3,512,703 214.9

Pantheon is now responding significantly faster following recent improvements to how we cache and retrieve data and handle API requests. We have also improved how Pantheon handles work that runs behind the scenes, such as file processing, status updates, and automated workflows. Users should see common tasks—viewing project details, accepting projects, loading dashboards feel more consistent and responsive throughout Pantheon.
We will continue to monitor and identify additional areas of improve Pantheon's performance. If you experience any specific issues, please submit a ticket with details (screenshots, time stamps, steps to reproduce, etc.) and we will investigate.

The Pantheon performance initiative is ongoing, a number of areas for performance and stability improvements have been identified. We will update again when the improvements have been implemented and validated.

We are continuing to work on a fix for this issue.

The root cause has been identified, and a fix is under way. We will update as soon as we have more information.

We are currently investigating reports of slow performance with Relay.