Need more concurrency? Need to increase rate limits?Please contact us for assistance, letting us know about your use case and requirements.
Concurrency
| Description | Limit |
|---|---|
| Concurrent image generation job | 10 (Default) |
| Concurrent model training generation job | 5 (Default) |
Note:
- Default concurrency can be increased on a custom API plan.
API Rate Limits
| Request Type | Endpoint | Limit |
|---|---|---|
| All request types | All | 2000 per minute (Default) |
| Create a Generation of Images requests | /v1/generations | 100 per minute (Default) |
| Create using Universal Upscaler requests | /v1/variations/universal-upscaler | 100 per minute (Default) |
| Create unzoom requests | /v1/variations/unzoom | 100 per minute (Default) |
| Create upscale requests | /v1/variations/upscale | 100 per minute (Default) |
| Create no background requests | /v1/variations/nobg | 100 per minute (Default) |
| Create Texture Generation requests | /v1/generations-texture | 100 per minute (Default) |
Note:
- 1 concurrency comes with 10 create a generation of images requests
Queue Limit
| Description | Limit |
|---|---|
| Image generation with status pending / queued | 200 (Default) |
| Upscaling with status pending / queued | 100 (Default) |
VEO3 Video generation with status pending / queued | 8 (Default) |
VEO3FAST Video generation with status pending / queued | 1 (Default) |
Note:
- 1 concurrency comes with 20 image generation with status pending