Generate File
https://api.youvico.com/api/projects/:id/files/upload.generate- MCP Tool
- Not Supported
- Rate Limit
- Tier 2
- Scope
- ai:generate
Generates a file with AI in the generation mode you choose.
AUTOanalyzes your prompt and attachments to plan the generation, and keeps any output settings you send inconfigs.ADVANCEDgenerates from your prompt andconfigsas sent. Put every attachment in an attachment slot ofconfigs.
If ADVANCED cannot use an attachment under the model's rules, the error response gives that attachment's ID in error.details.attachment.id and the reason in error.details.reason.
Available only in projects of the workspace your API key or connection belongs to, not in projects another workspace has linked to it.
Note
To check the generation result, poll Get File until status is UPLOADED (succeeded) or FAILED (failed).
Example
curl -X POST 'https://api.youvico.com/api/projects/bdbff5de-96d7-468f-9db0-85fe28bd6b62/files/upload.generate' \
-H 'Authorization: Bearer YOUR_API_KEY' \
-H 'Content-Type: application/json' \
-d '{ "mode": "ADVANCED", "output": { "type": "VIDEO", "model": "kling-3.0" }, "prompt": "The camera slowly pushes in as the sneaker turns on a turntable.", "folder": { "id": "DX1234567890ABCD" }, "attachments": [{ "id": "GA1234567890ABCD" }], "configs": { "start": "GA1234567890ABCD", "durationSeconds": 10 } }'
{
"data": {
"id": "FXABCDEFGHIJKLMN"
}
}
{
"data": {
"id": "FXABCDEFGHIJKLMN"
}
}
Request
Path
idstringrequiredBody
modestringrequiredPossible values: AUTOADVANCED
Example: ADVANCED
outputobjectrequiredoutput.typestringrequiredPossible values: PDFMARKDOWNIMAGEVIDEO3DMUSIC
Example: VIDEO
output.modelstringrequiredExample: kling-3.0
promptstringrequired1-20000 characters
Example: The camera slowly pushes in as the sneaker turns on a turntable.
folderobject?requiredfolder.idstringrequiredExample: DX1234567890ABCD
attachmentsarrayrequired0-50 items, unique IDs
attachments[].idstringrequiredExample: GA1234567890ABCD
configsobjectNote
In ADVANCED, every ID in attachments must appear in an attachment slot of configs, and the slots cannot name any other ID.
Response
Returns 202 Accepted.
dataobjectdata.idstringExample: FXABCDEFGHIJKLMN
Models
List Models returns each model's settings schema. Document models (PDF, MARKDOWN) support ADVANCED only.
| Output type | Models |
|---|---|
PDF | claude-fable-5-1, claude-opus-5-5, claude-sonnet-5-5, claude-haiku-5-5, gpt-6-astra, gpt-6.1-sol, gpt-6-luna, gemini-4-argon, gemini-3.8-flash |
MARKDOWN | claude-fable-5-1, claude-opus-5-5, claude-sonnet-5-5, claude-haiku-5-5, gpt-6-astra, gpt-6.1-sol, gpt-6-luna, gemini-4-argon, gemini-3.8-flash |
IMAGE | gpt-image-2.5-flare, gpt-image-2.5-sunburst, gpt-image-2, gemini-nano-banana-2.1, gemini-3-pro-image, gemini-3.1-flash-image, gemini-3.1-flash-lite-image |
VIDEO | seedance-2.5, seedance-2.0, seedance-2.0-fast, seedance-2.0-mini, kling-3.0-omni, kling-3.0, kling-3.0-turbo, minimax-h3-max, minimax-h3, veo-3.1-generate-preview, veo-3.1-lite-generate-preview |
3D | meshy-7.1, hunyuan3d-3.1 |
MUSIC | lyria-3.5 |
Settings
Accepted values and defaults are in each model's schema. In AUTO, settings you omit are chosen by YouViCo. In ADVANCED, they use their defaults.
| Setting | Description |
|---|---|
effort | Thinking level. Higher levels think longer |
aspectRatio | Aspect ratio. adaptive follows the input image or video |
resolution | Output resolution |
quality | Image quality |
outputFormat | Image or audio file format |
transparent | Transparent background. Must be false when outputFormat is jpeg |
durationSeconds | Output length in seconds |
generateAudio | Whether to generate sound |
shouldTexture | Whether to apply textures |
MUSIC models take only the outputFormat setting. In ADVANCED, describe lyrics, length, tempo, and key in the prompt.
Attachment Slots
Only ADVANCED schemas have attachment slots. Each model's schema shows which slots it has and the rules between them.
| Slot | Description |
|---|---|
references | Reference images, videos, audio, or documents |
start, end | Start and end frame images |
edit | Video to edit. The output follows its length and aspect ratio |
extend | Video to extend by durationSeconds |
motion | Motion control. image and video are required |
images | Input images for 3D. hunyuan3d-3.1 takes an object keyed by view and requires front |
Some video models also accept negativePrompt, what to avoid in the output, and shots, a multi-shot list of prompt and seconds that add up to the output length.