Model Studio Wan Video

---

name: alicloud-ai-video-wan-video

description: Generate videos with Model Studio DashScope SDK using Wan i2v models (wan2.6-i2v-flash, wan2.6-i2v, wan2.6-i2v-us). Use when implementing or documenting video.generate requests/responses, mapping prompt/negative_prompt/duration/fps/size/seed/reference_image/motion_strength, or integrating video generation into the video-agent pipeline.

version: 1.0.0

---

Category: provider

# Model Studio Wan Video

Validation

mkdir -p output/alicloud-ai-video-wan-video
python -m py_compile skills/ai/video/alicloud-ai-video-wan-video/scripts/generate_video.py && echo "py_compile_ok" > output/alicloud-ai-video-wan-video/validate.txt

Pass criteria: command exits 0 and `output/alicloud-ai-video-wan-video/validate.txt` is generated.

Output And Evidence

Provide consistent video generation behavior for the video-agent pipeline by standardizing `video.generate` inputs/outputs and using DashScope SDK (Python) with the exact model name.

Critical model names

Use one of these exact model strings:

Prerequisites

python3 -m venv .venv
. .venv/bin/activate
python -m pip install dashscope

Normalized interface (video.generate)

Request

Response

Quick start (Python + DashScope SDK)

Video generation is usually asynchronous. Expect a task ID and poll until completion.

Note: Wan i2v models require an input image; pure t2v models can omit `reference_image`.

import os
from dashscope import VideoSynthesis

# Prefer env var for auth: export DASHSCOPE_API_KEY=...
# Or use ~/.alibabacloud/credentials with dashscope_api_key under [default].

def generate_video(req: dict) -> dict:
    payload = {
        "model": req.get("model", "wan2.6-i2v-flash"),
        "prompt": req["prompt"],
        "negative_prompt": req.get("negative_prompt"),
        "duration": req.get("duration", 4),
        "fps": req.get("fps", 24),
        "size": req.get("size", "1280*720"),
        "seed": req.get("seed"),
        "motion_strength": req.get("motion_strength"),
        "api_key": os.getenv("DASHSCOPE_API_KEY"),
    }

    if req.get("reference_image"):
        # DashScope expects img_url for i2v models; local files are auto-uploaded.
        payload["img_url"] = req["reference_image"]

    response = VideoSynthesis.call(**payload)

    # Some SDK versions require polling for the final result.
    # If a task_id is returned, poll until status is SUCCEEDED.
    result = response.output.get("results", [None])[0]

    return {
        "video_url": None if not result else result.get("url"),
        "duration": response.output.get("duration"),
        "fps": response.output.get("fps"),
        "seed": response.output.get("seed"),
    }

Async handling (polling)

import os
from dashscope import VideoSynthesis

task = VideoSynthesis.async_call(
    model=req.get("model", "wan2.6-i2v-flash"),
    prompt=req["prompt"],
    img_url=req["reference_image"],
    duration=req.get("duration", 4),
    fps=req.get("fps", 24),
    size=req.get("size", "1280*720"),
    api_key=os.getenv("DASHSCOPE_API_KEY"),
)

final = VideoSynthesis.wait(task)
video_url = final.output.get("video_url")

Operational guidance

Size notes

Output location

Anti-patterns

Workflow

1) Confirm user intent, region, identifiers, and whether the operation is read-only or mutating.

2) Run one minimal read-only query first to verify connectivity and permissions.

3) Execute the target operation with explicit parameters and bounded scope.

4) Verify results and save output/evidence files.

References