#videopose
Upload a photo of a person and match it with a video pose (official template, your custom video pose, or directly specify the first frame picture pose) to generate a pose video of the character. Tasks are executed asynchronously and usually take one to several minutes.
Create task
POST /v1/videos/pose| Parameters | Type | Required | Description |
|---|---|---|---|
template_id | string | Choose one of two | Template ID (see "Template List"), or your own custom video pose ID (starting with uvpose_, see "Custom Video Pose") |
imgpose_id | string | Choose one of two | Directly specify the first frame picture pose ID (official pose_ or your upose_), one-time, no saving. Choose one between template_id and template_id for both transmissions |
work_type | int | No | Video engine, only effective when using imgpose_id: 1=Wan2.2 (fixed 5 seconds), 2=LTX (default, adjustable duration) |
prompt | string | No | Video motion/scene prompt word, only effective when using imgpose_id |
image | string | Yes | Person photo: image URL, or base64 (can be prefixed with data:image/jpeg;base64,). Max 10MB |
duration | integer | No | Video duration (seconds). It only takes effect for the adjustable duration template (duration_adjustable: true), and the range is min_duration–max_duration of the template; the fixed duration template will ignore this parameter. See Billing and Points |
webhook_url | string | No | Push address after the task ends |
The price changes with the template and duration. The cost field of the task object is the points deducted this time. For details, see Billing and Points.
Request header Idempotency-Key (optional): Your business ID, which is not repeated within the same function, will be returned unchanged in the result.
Photo requirements:
- Frontal face with clear facial features. It is best to have only one person in the picture.
- If no clear face is detected, or a minor is suspected,
422 input_rejectedwill be returned.
Example
curl -X POST https://api.modelgo.moneygo.ai/v1/videos/pose \
-H "Authorization: Bearer sk-your-key" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: order_10086" \
-d '{
"template_id": "video_dance",
"image": "https://example.com/face.jpg",
"webhook_url": "https://your-server.com/webhook"
}'import base64
import requests
with open("face.jpg", "rb") as f:
image = "data:image/jpeg;base64," + base64.b64encode(f.read()).decode()
res = requests.post(
"https://api.modelgo.moneygo.ai/v1/videos/pose",
headers={"Authorization": "Bearer sk-your-key", "Idempotency-Key": "order_10086"},
json={"template_id": "video_dance", "image": image},
timeout=60,
)
print(res.json())import { readFile } from 'node:fs/promises'
const image = 'data:image/jpeg;base64,' + (await readFile('face.jpg')).toString('base64')
const res = await fetch('https://api.modelgo.moneygo.ai/v1/videos/pose', {
method: 'POST',
headers: { 'Authorization': 'Bearer sk-your-key', 'Content-Type': 'application/json', 'Idempotency-Key': 'order_10086' },
body: JSON.stringify({ template_id: 'video_dance', image }),
})
console.log(await res.json())Return
{
"id": "task_3kP9xQ2mZ7aB1cD4eF6gH8jK",
"object": "task",
"capability": "video.pose",
"status": "queued",
"template_id": "video_dance",
"cost": 140,
"output": null,
"error": null,
"created_at": "2026-09-17T08:00:00.000Z",
"started_at": null,
"finished_at": null
}Then get the results via a webhook callback or task query. When successful, output.images is the result video address (.mp4).
TIP
The result field name follows output.images, which is consistent with the posture image; in the video mode, the array contains the video address.
Template list
GET /v1/templates?capability=video.pose&page=1&page_size=20Optional query parameter duration_adjustable: true returns only adjustable-duration templates (LTX), false returns only fixed-duration templates (Wan2.2), and omitting it returns all templates. Only true and false are accepted. This filter only applies to capability=video.pose. Filtering happens before pagination, so total is the number of matching templates. Without pagination parameters, the response contains the full filtered list.
GET /v1/templates?capability=video.pose&duration_adjustable=true&page=1&page_size=20Pass page or page_size to paginate. See Template pagination for parameters and response fields.
{
"data": [
{
"id": "video_dance",
"capability": "video.pose",
"name": "Dance",
"description": "",
"price": 100,
"duration_adjustable": true,
"min_duration": 5,
"max_duration": 15,
"duration_prices": { "5": 100, "6": 120, "7": 130, "8": 140 },
"cover_url": "https://cdn.example.com/covers/dance.jpg"
}
],
"total": 1,
"page": 1,
"page_size": 20
}Templates will increase or decrease from time to time. It is recommended to synchronize regularly and do not hard-code the template ID. For price, duration and the pricing rules, see Billing and Points.
Custom video pose
You can use a picture pose as the first frame and use the video engine to generate your own video pose - in two ways:
- One-time: Directly pass
imgpose_id(+ optionalwork_type/prompt/duration) when creating a task without saving. - Save and reuse: Save the "first frame picture pose + video settings" as a private video pose, get the
uvpose_ID, and use it astemplate_idlater.
The first frame of imgpose_id can be the official picture template (pose_) or your own [custom picture pose] (/pose#custom pose) (upose_). When generating, first use it to generate the first frame pose of your photo, and then generate a video based on it. The billing is consistent with the official video template (default price per engine, LTX per second). Custom video poses not include LoRA, the movements are mainly driven by the engine and prompt words.
| Method | Description |
|---|---|
POST /v1/video-poses | Created. Body: { name, imgpose_id, work_type?, prompt?, work_duration?, cover_url? }. Return containing id (uvpose_...) |
GET /v1/video-poses | My custom video pose list · Template pagination |
PATCH /v1/video-poses/{id} | Modify |
DELETE /v1/video-poses/{id} | Delete |
# 1. Save a video pose (official image pose `pose_office` as the first frame, LTX, 5 seconds by default)
curl -X POST https://api.modelgo.moneygo.ai/v1/video-poses \
-H "Authorization: Bearer sk-your-key" -H "Content-Type: application/json" \
-d '{ "name": "Office turn", "imgpose_id": "pose_office", "work_type": 2, "prompt": "Turn around slowly", "work_duration": 5 }'
# → { "id": "uvpose_xxx", ... }
# 2. Generate with it (same as an official template)
curl -X POST https://api.modelgo.moneygo.ai/v1/videos/pose \
-H "Authorization: Bearer sk-your-key" -H "Content-Type: application/json" \
-d '{ "template_id": "uvpose_xxx", "image": "https://example.com/face.jpg", "duration": 5 }'