API docs / 3D models

hunyuan3d-2mv API

Generates 3D assets from multiple view images with multiview controlled shape generation. Takes front, back, left, and right view images as input and produces textured 3D models in GLB, OBJ, PLY, or STL format. Finetuned from Hunyuan3D-2 specifically to support multiview-controlled 3D generation, allowing precise control over the generated 3D shape through multiple perspective images. Includes automatic background removal and mesh simplification options for optimized 3D models.

Model id
tencent/hunyuan3d-2mv
Provider
tencent
Output
3D
Price
From 42 credits per run, the same as on voor.ai. Options such as length or resolution can change it.

Request

POST https://voor.ai/v1/generations with "model": "tencent/hunyuan3d-2mv". Poll GET /v1/tasks/{id} or receive the signed webhook. See the overview for keys, idempotency, webhooks and errors.

curl https://voor.ai/v1/generations \
  -H "Authorization: Bearer $VOOR_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
  "model": "tencent/hunyuan3d-2mv",
  "input": {
    "front_image": [
      "https://example.com/input.png"
    ]
  },
  "callback_url": "https://example.com/webhooks/voor"
}'

# Then poll the task (or wait for the webhook):
curl https://voor.ai/v1/tasks/TASK_ID \
  -H "Authorization: Bearer $VOOR_API_KEY"

Input parameters

NameTypeDetails
seednumber

Random seed

Default: 1234

guidancenumber

Guidance scale

Default: 5

stepsnumber

Number of inference steps

Default: 30

Min 1 · Max 100

output_formatstring

Default: glb

One of: glb, obj, ply, stl

back_imagearray of URLs

Back view image

Up to 1 files · Accepts image/*

front_imagerequiredarray of URLs

Front view image

Up to 1 files · Accepts image/*

left_imagearray of URLs

Left view image

Up to 1 files · Accepts image/*

num_chunksnumber

Number of chunks

Default: 200000

Min 1000 · Max 5000000

octree_resolutionnumber

Octree resolution

Default: 256

Min 16 · Max 512

randomize_seedboolean

Randomize seed

Default: true

remove_backgroundboolean

Remove image background

Default: true

right_imagearray of URLs

Right view image

Up to 1 files · Accepts image/*

target_face_numnumber

Target number of faces for mesh simplification

Default: 10000

Min 100 · Max 1000000