FASHN - Notice history

We have identified delayed response times from a downstream service as the cause of elevated generation latency and intermittent timeout errors. We are working to increase capacity and return the service to normal operation.

It is recommend to temporarily use "Fast" generation mode where possible for a significantly reduced error rate.

This issue doesn't affect tryon-v1.5 and tryon-v1.6 API endpoints.

We are currently investigating this incident.

Try-On v1.5 & v1.6 are not affected.

We have identified the root cause and are currently working on a fix.

The pipeline running "Face Reference" is experiencing technical issues; Generations with "Face Reference" can sometimes timeout.

Degraded availability in services supporting the Try-On v1.6 pipeline are causing longer than usual processing times. We are working on adding alternative providers as fallbacks to fix the issue and also make it more reliable for the future.

We're upgrading pieces of our server infrastructure, downtime expected for a few minutes only.

A technical issue is causing tryon-v1.6 API requests to fail if you rely on category: auto and garment_photo_type: auto .

The team is working on a fix.

Our app and developer API are experiencing intermittent failures due to technical issues in a cloud provider we rely on.

Tools that have the Generation Model option of 'Quality' will have a longer than usual processing time for this setting.

Consider temporarily switching to Fast or Balanced for a better experience.

For API users:

generation_mode: fast or generation_mode: balanced in supported endpoints.