curl --request POST \
--url https://api.poyo.ai/api/generate/submit \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "meshy-7.1-text-to-3d",
"input": {
"prompt": "A ceramic robot with a clean silhouette.",
"enable_rigging": true,
"enable_animation": true
}
}
'
{
"code": 200,
"data": {
"task_id": "task-unified-1757165031-uyujaw3d",
"status": "not_started",
"created_time": "2026-10-13T10:30:00+08:00"
}
}Meshy
Meshy 7.1 3D
Generate 3D models from text, one image, or 1–4 images, with optional textures, rigging, and animation.
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": "meshy-7.1-text-to-3d",
"input": {
"prompt": "A ceramic robot with a clean silhouette.",
"enable_rigging": true,
"enable_animation": true
}
}
'
{
"code": 200,
"data": {
"task_id": "task-unified-1757165031-uyujaw3d",
"status": "not_started",
"created_time": "2026-10-13T10:30:00+08:00"
}
}Submission returns a
Image inputs and
Optional features can change the task price; inspect the live Catalog before submitting.
task_id. Use Query Task Status to track completion, or supply callback_url for completion and failure notifications.
Meshy 7.1 supports three generation modes through separate model IDs.
Available models
| Model | Required input |
|---|---|
meshy-7.1-text-to-3d | prompt: 1–600 characters, containing non-whitespace text |
meshy-7.1-image-to-3d | image_urls: exactly one image |
meshy-7.1-multi-image-to-3d | image_urls: 1–4 images of the same object from different angles |
texture_image_url accept public HTTP(S) image URLs or image data URIs.
Parameters and combinations
All parameters appear in one list. Each description identifies model-specific requirements, supported values, and defaults.| Parameter | Description |
|---|---|
model | Required. Select one of the three model IDs listed above. |
input | Required. Object containing the generation parameters below. Fields marked as required for a model must be supplied when that model is selected. |
callback_url | Optional completion or failure callback URL. |
input.prompt | Required only for meshy-7.1-text-to-3d; omit for image models. Use 1–600 characters containing non-whitespace text. |
input.image_urls | Required for image models; omit for meshy-7.1-text-to-3d. meshy-7.1-image-to-3d requires exactly 1 image; meshy-7.1-multi-image-to-3d accepts 1–4 images of the same object from different angles. Each item accepts a public HTTP(S) image URL or image data URI. |
input.mode | Only for meshy-7.1-text-to-3d. preview returns geometry only; full includes textures. Default: full. |
input.model_type | Only for meshy-7.1-text-to-3d and meshy-7.1-image-to-3d; omit for meshy-7.1-multi-image-to-3d. Accepts standard (default) or smart-topology. Lowpoly is not supported. |
input.geometry_resolution | meshy-7.1-text-to-3d and meshy-7.1-image-to-3d accept standard, 2k, or 4k only when model_type is standard (the default); omit for smart-topology. meshy-7.1-multi-image-to-3d accepts only standard or 2k and does not support 4k. |
input.topology | Output mesh topology for all models: triangle (default) or quad. |
input.target_polycount | Target polygons for all models: 100–300000; default 30000. For meshy-7.1-text-to-3d and meshy-7.1-image-to-3d with model_type=smart-topology: 100–15000, default 4000 when omitted. |
input.should_remesh | Enable remeshing for all models. Default: true. |
input.symmetry_mode | Mesh symmetry behavior for all models: off, auto (default), or on. |
input.should_texture | Only for meshy-7.1-image-to-3d and meshy-7.1-multi-image-to-3d; omit for meshy-7.1-text-to-3d. Generate textures; default true. For multi-image generation, must remain true when enable_pbr=true or texture guidance is supplied. |
input.enable_pbr | Generate PBR texture maps for all models. Default: false. For meshy-7.1-multi-image-to-3d, requires should_texture=true. |
input.pose_mode | Character pose for all models: a-pose, t-pose, or an empty string (default) for no specific pose. |
input.seed | Only for meshy-7.1-text-to-3d. Integer random seed; omit to select automatically. |
input.enable_prompt_expansion | Only for meshy-7.1-text-to-3d. Expand the prompt before generation. Default: false. |
input.texture_prompt | Optional texture guidance for all models, at most 600 characters. For meshy-7.1-multi-image-to-3d, requires should_texture=true. |
input.texture_image_url | Optional texture guidance for all models; accepts a public HTTP(S) image URL or image data URI. For meshy-7.1-multi-image-to-3d, requires should_texture=true. |
input.enable_rigging | Rig the generated humanoid character for all models. Default: false. Must be true when enable_animation=true. |
input.rigging_height_meters | Character height in meters for all models when rigging is enabled; must be greater than zero. Default: 1.7. |
input.enable_animation | Apply an animation for all models; requires enable_rigging=true. Default: false. |
input.animation_action_id | Animation preset ID for all models: integer from 0 to 696. Default: 92. |
input.enable_safety_checker | Enable input safety checking for all models. Default: true. Disabling requires account authorization when applicable. |
Examples
Text to 3D with rigging and animation
{
"model": "meshy-7.1-text-to-3d",
"input": {
"prompt": "A ceramic robot with a clean silhouette.",
"enable_rigging": true,
"enable_animation": true
}
}
Text to 3D with Smart Topology preview
{
"model": "meshy-7.1-text-to-3d",
"input": {
"prompt": "A ceramic robot.",
"mode": "preview",
"model_type": "smart-topology"
}
}
Single image to 3D with rigging and animation
{
"model": "meshy-7.1-image-to-3d",
"input": {
"image_urls": [
"https://example.com/front.png"
],
"enable_rigging": true,
"enable_animation": true
}
}
Multiple images to 3D with rigging and animation
{
"model": "meshy-7.1-multi-image-to-3d",
"input": {
"image_urls": [
"https://example.com/front.png",
"https://example.com/back.png"
],
"enable_rigging": true,
"enable_animation": true
}
}
Task results
The query response uses the common task structure:task_id, status, progress, files, and error_message. A finished task has status="finished"; each result file has file_url and file_type. Preserve the task ID while polling.
3D assets appear as file_type="3d", with thumbnails and textures as images. Rigged meshes and animation assets are returned when requested and available.Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Body
application/json
Select meshy-7.1-text-to-3d for text, meshy-7.1-image-to-3d for exactly one image, or meshy-7.1-multi-image-to-3d for 1–4 images.
Available options:
meshy-7.1-text-to-3d, meshy-7.1-image-to-3d, meshy-7.1-multi-image-to-3d Unified parameter list. Required fields, supported values, and defaults depend on model as described for each parameter.
Show child attributes
Show child attributes
Optional completion or failure callback URL.