GET
/v1/schemas/fal/video/fal-ai/bernini-r/edit-video
Accept:
| field | recorded source |
|---|---|
| schema | Source linked above. upstream-spec Source traceb9ed491719a671aa65b3007718e63b538aa59cbde29c84001bb46e2eacee3aa3{
"fetchedAt": 1791522002,
"extractorVersion": "1",
"verifiedAt": null
} |
Request body
11 properties · 2 required · 0 $defs| property | type | description | constraints |
|---|---|---|---|
| prompt* | string | Text prompt or editing instruction. | |
| video_url* | string | Source video to edit. | |
| frames_per_second | integer | Output frames per second. | 4 ≤ n ≤ 30 · default: 16 |
| num_frames | integer | Number of frames. Snapped internally to 4k+1. | 5 ≤ n ≤ 121 · default: 81 |
| seed | integer | Seed for reproducibility. Random when omitted. | nullable |
| num_inference_steps | integer | Number of denoising steps. | 1 ≤ n ≤ 50 · default: 30 |
| enable_safety_checker | boolean | Authorized callers may set false to request relaxed screening. | default: true |
| negative_prompt | string | Negative prompt. Defaults to the standard Wan2.2 negative prompt. | nullable |
| max_image_size | integer | Long-edge size (px). Sets the output size for generation tasks and caps source/reference media for editing tasks. | 256 ≤ n ≤ 1280 · default: 848 |
| acceleration | enum | Acceleration level. 'regular' enables MagCache step-skipping for faster generation at a small quality cost; 'none' (default) runs the full denoising schedule. | none | regular · default: "none" |
| enable_prompt_expansion | boolean | Rewrite the prompt with an LLM before generation. The model is tuned on enhanced prompts; off by default. | default: false |
Validate a payload
before you spend tokensshell
$ curl -X POST https://modelschemas.com/v1/validate \
-d '{"provider":"fal","endpointId":"fal-ai/bernini-r/edit-video","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}