curl --request POST \
--url https://api.poyo.ai/api/generate/submit \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "tripo3d-h3.1-text-to-3d",
"input": {
"prompt": "A japanese samurai kabuto helmet with gold dragon ornament and red cord lacing",
"texture": true,
"texture_quality": "standard",
"geometry_quality": "standard",
"quad": false
}
}
'{
"code": 200,
"data": {
"task_id": "<string>",
"status": "not_started",
"created_time": "2023-11-07T05:31:56Z"
}
}Tripo3D
Tripo3D H3.1 API
High-detail Tripo3D H3.1 text-to-3D, image-to-3D, and multiview-to-3D generation
POST
/
api
/
generate
/
submit
curl --request POST \
--url https://api.poyo.ai/api/generate/submit \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "tripo3d-h3.1-text-to-3d",
"input": {
"prompt": "A japanese samurai kabuto helmet with gold dragon ornament and red cord lacing",
"texture": true,
"texture_quality": "standard",
"geometry_quality": "standard",
"quad": false
}
}
'{
"code": 200,
"data": {
"task_id": "<string>",
"status": "not_started",
"created_time": "2023-11-07T05:31:56Z"
}
}- After submission, a
task_idwill be returned. If you provided acallback_url, PoYo sends a POST request when the task becomesfinishedorfailed. - You can retrieve results through the unified Query Task Status endpoint.
Tripo3D H3.1 API
Tripo3D H3.1 supports high-detail text-to-3D, single image-to-3D, and multiview-to-3D generation through three public model IDs.Available Models
- tripo3d-h3.1-text-to-3d - Generate a high-detail 3D model from a prompt
- tripo3d-h3.1-image-to-3d - Generate a 3D model from one object image
- tripo3d-h3.1-multiview-to-3d - Generate a 3D model from 2-4 views of the same object
Required Parameters
- model:
tripo3d-h3.1-text-to-3d,tripo3d-h3.1-image-to-3d, ortripo3d-h3.1-multiview-to-3d - input.prompt: Required for
tripo3d-h3.1-text-to-3d - input.image_urls: Required for image models. Use exactly one item for image-to-3D; use 2-4 items for multiview-to-3D
Optional Parameters
- negative_prompt: Text-to-3D only. Features to avoid
- face_limit: Target face count, from
1000to2000000 - texture: Whether to generate textures. Default is
true - pbr: Whether to generate PBR materials when textures are enabled
- model_seed, image_seed, texture_seed: Optional integer seeds
- texture_quality:
standardordetailed. Default isstandard - geometry_quality:
standardordetailed. Default isstandard - auto_size: Whether to auto-scale the model
- quad: Whether to generate quad mesh topology
- callback_url: Optional webhook URL for task completion notification
Pricing
- Text-to-3D and multiview-to-3D: 15 credits without textures, 30 credits with standard textures, or 45 credits with detailed textures
- Image-to-3D: 30 credits without textures, 45 credits with standard textures, or 60 credits with detailed textures
- Detailed geometry adds 30 credits
- Quad mesh adds 7.5 credits
Output
The task result is returned as a file list. Available items may include:model_glbthumbnail- entries from
model_urls, such asglb,base_model, andpbr_model, when available
label, file_url, file_type, format, content_type, file_name, and file_size when available.
Notes
- Multiview inputs should show the same object from different angles.
- The API returns asynchronous task status. Poll the unified task status endpoint or use
callback_url.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Body
application/json