Reference
API reference
Complete input parameter surface for GPT Image 2. The canonical schema lives at the model's llms.txt; the table below is derived from it.
Endpoint
Submit a job to the queue endpoint:
POST https://queue.fal.run/fal-ai/gpt-image-2Input parameters
| Key | Type | Default | Allowed / notes |
|---|---|---|---|
| promptrequired | string | — | Free text describing the output. |
| size | enum | 1024x1024 | 1024x1024 | 1536x1024 | 1024x1536 | 2048x2048 | 2048x1152 | 1152x2048 |
| quality | enum | medium | low | medium | high |
| num_images | enum | 1 | 1 | 2 | 4 |
| safety_tolerance | number | 2 | Sent on every request unless overridden. |
Canonical schema: fal.ai/models/fal-ai/gpt-image-2/llms.txt. Anything new shipped by the model vendor appears there first.
Worked example
Minimal request using the defaults above. Swap the prompt for your own; everything else stays optional.
BASH
1curl -X POST "https://queue.fal.run/fal-ai/gpt-image-2" \2 -H "Authorization: Key $FAL_KEY" \3 -H "Content-Type: application/json" \4 -d '{"prompt":"A dense East-Asian bodega storefront at dusk, hand-painted signs with the exa...","safety_tolerance":2,"size":"1024x1024","quality":"medium","num_images":1}'