Wan 3.0 Prime Reference-to-Video Spicy

Wan 3.0 Prime Reference-to-Video Spicy is Alibaba's flagship uncensored high-motion video model engineered for native 4K visual synthesis. Delivering native 4K output at 60 fps, it processes up to 8 reference images or 2 reference videos simultaneously to preserve character identity and material fidelity across complex sequences. With unrestricted generation capabilities, it synthesizes explosive physical movements, dramatic camera trajectories, and physical interactions—powering high-energy cinematic pre-visualization, high-impact commercial ads, AAA gaming assets, and action pipelines.

Wan 3.0 Prime Reference-to-Video Spicy

Wan 3.0 Prime Reference-to-Video Spicy is an ultra-fast, high-quality AI reference-to-video-global model from Alibaba’s Tongyi Wanxiang series. Combining the ultra-fast rendering capabilities of the Prime architecture with Wan 3.0’s multimodal foundation, the model supports single-image first-frame guidance and smooth transitions between first and last frames. It can directly generate 1080p HD videos up to 30 seconds long. The model natively integrates joint audio-video generation, including ambient sound effects and multilingual character lip synchronization. In addition to significantly reducing video rendering wait times, it provides highly accurate motion physics simulation, camera movement control, and subject consistency. It is widely applicable to dynamic e-commerce product presentations, film and television visual effects, rapid short-drama iteration, and commercial advertising production.

Base URL

https://api.icreat.ai

Authentication

All API requests require authentication using an API key. You can obtain an API key from the console.

export ICREAT_API_KEY="your-api-key-here"

HTTP Request Headers

import os

API_KEY = os.environ.get("ICREAT_API_KEY")
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer " + API_KEY,
}

Keep Your API Key Secure

Never expose your API key in client-side code or public repositories. Use environment variables or a backend proxy.

Code Examples

Image/video generation uses a three-step asynchronous workflow: first submit a task to obtain a task_id, then poll the task status, and finally retrieve the generated result when the status is SUCCEEDED. The following examples use the same task_id across all three steps.

1. Submit a Task

Send a generation request to the submission endpoint.

POST/v1/task/submit/aliyun/wan3-0-prime/reference-to-video-global

curl --fail-with-body --connect-timeout 10 --max-time 60 \
  -X POST https://api.icreat.ai/v1/task/submit/aliyun/wan3-0-prime/reference-to-video-global \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "x-icreat-ai-group: wan-spicy" \
  -H "Content-Type: application/json" \
  -d '{
    "input": {
      "prompt": "A running kitten",
      "negative_prompt": "The kitten should not run in a straight line",
      "media": [
        {"type": "reference_image", "url": "..."},
        {"type": "reference_video", "url": "..."},
        {"type": "reference_audio", "url": "..."}
      ]
    },
    "parameters": {
      "resolution": "720P",
      "duration": 5,
      "watermark": false
    }
  }'

2. Poll the Status

Use the task_id to query the task’s progress. The response body contains only the status field.

POST/v1/task/query-status

curl --connect-timeout 10 --max-time 60 \
  -X POST "https://api.icreat.ai/v1/task/query-status" \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "task_id": "task-xxx"
  }'

3. Retrieve the Result

Retrieve the final output after the task succeeds.

POST/v1/task/get-result

curl --connect-timeout 10 --max-time 60 \
  -X POST "https://api.icreat.ai/v1/task/get-result" \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "task_id": "task-xxx"
  }'

Input Parameters

Submit Task — Input Parameters

The following parameters are accepted in the task submission request body.

Total: 2; Required: 2; Optional: 0

inputobjectrequired

The input content for the generation task.

parametersobjectrequired

Generation parameters.

Poll Status — Input Parameters

Total: 1; Required: 1; Optional: 0

task_idstringrequired

The task ID returned by the task submission endpoint.

Get Result — Input Parameters

Total: 1; Required: 1; Optional: 0

task_idstringrequired

The task ID returned by the task submission endpoint.

Output Parameters

Submit Task — Output Parameters

Total: 1

task_idstring

The task ID returned by the task submission endpoint.

Poll Status — Output Parameters

Total: 1

statusstring

The current task status. The result can be retrieved when the status is SUCCEEDED.

Get Result — Output Parameters

The response body is an array of resource objects (array[object]).

Total: 3

typestring

The resource type. For video models, this is Video.

urlstring

The preview/access URL.

download_urlstring

The download URL.

LLM-Friendly Prompt

The following is an LLM-friendly Markdown prompt that can be copied into AI assistants such as Cursor or ChatGPT to help the AI understand the model’s API integration method, invocation workflow, and key parameters. Click the “Copy LLM Prompt” button or copy the full text from the code block below.

# aliyun/wan3-0-prime/reference-to-video-global

> Wan 3.0 Prime Reference-to-Video is an ultra-fast, high-quality AI image-to-video model from Alibaba’s Tongyi Wanxiang series.

## Overview

Submit a generation request through iCreat’s three-step asynchronous task API, poll the task status, and retrieve the video resource URL after the task succeeds.

## API Information

- **Base URL**: `https://api.icreat.ai`
- **Submit endpoint (POST)**: `/v1/task/submit/aliyun/wan3-0-prime/reference-to-video-global`
- **Poll endpoint (POST)**: `/v1/task/query-status`
- **Get result endpoint (POST)**: `/v1/task/get-result`
- **Model ID**: `aliyun/wan3-0-prime/reference-to-video-global`
- **Authentication**: `Authorization: Bearer ${ICREAT_API_KEY}`

## Invocation Workflow

1. **Submit**: Send a POST request to the submit endpoint using the body described under Input Notes. The response is `{ "task_id": "..." }`.
2. **Poll**: Send a POST request to `/v1/task/query-status` with the body `{ "task_id": "..." }`. The response contains **only** `{ "status": "SUBMITTED|SUCCEEDED|FAILED" }`.
3. **Get result**: When `status` is `SUCCEEDED`, send a POST request to `/v1/task/get-result`. The response is a top-level array. When `type` is `Video`, read `url` or `download_url`.

### Input Notes

- Top-level request body: `input` (object) and `parameters` (object). These are sibling fields.
- `input.prompt` (required): A prompt describing the video to generate.
- `input.negative_prompt` (optional): A description of content that should not appear in the video.
- `input.media` (optional): A list of reference media used for reference-to-video-global generation.
- `parameters.resolution` (required): The video resolution.
- `parameters.ratio` (required): The video aspect ratio.
- `parameters.duration` (required): The video duration in seconds.
- `parameters.watermark` (optional): Whether to add a watermark.

### Output Notes

- Polling: Check only `status`.
- Result: A top-level array in the form `[{ "type": "Video", "url": "...", "download_url": "..." }]`.

## Important Notes

- The same `task_id` must be used across all three steps. Do not skip polling.
- `FAILED` is a terminal status. Check the request parameters or reference content to troubleshoot the failure.
- In the submit request body, `input` and `parameters` are sibling fields. Do not incorrectly nest one inside the other.