Multi-shot narrative
T2V and I2V can turn a prompt into connected shots while preserving the main subject across transitions.
Wan 2.6 model guide
Wan 2.6 is Alibaba Cloud's hosted multimodal generation family for text-to-video, image-to-video and character reference workflows, with native audio, multi-shot narration and cost-focused Flash endpoints.
2–15sOutput duration
1080PMaximum resolution
30 fpsFixed frame rate
A/VSynchronized audio
InputsText · image · audio · video reference
Model familyWan 2.6
Output720P / 1080P · 30 fps · MP4
What this model family is
Released in December 2025, Wan 2.6 moved the hosted Wan API from fixed short silent clips toward 2–15 second audiovisual production. It is a group of Model Studio endpoints rather than one downloadable model: the right choice depends on whether you start with text, a first frame or reusable character references.
Production capabilities
The useful distinction is not a benchmark score, but which input assets and production controls each endpoint accepts.
Native audiovisual output
Generate automatic sound or supply an audio file so dialogue, ambience and movement share one timeline.
Character reference video
R2V uses reference images or videos to carry one or more characters into a newly described performance.
Flash cost controls
Flash I2V and R2V offer audio or silent modes at lower list prices for drafts and batch exploration.
Models and endpoints
These are hosted Model Studio IDs. Standard, Flash and US-scoped IDs have different pricing, duration rules and deployment availability.
| Model ID | Task | Inputs | Output | Use it when |
|---|---|---|---|---|
wan2.6-t2v | Text to video | Text + optional audio | 720P / 1080P · 2–15s · 30 fps · MP4 | General 15-second audiovisual generation |
wan2.6-i2v | Image to video | Text + first image + optional audio | 720P / 1080P · 2–15s · 30 fps · MP4 | Animate a controlled opening frame |
wan2.6-i2v-flash | Fast image to video | Text + first image + optional audio | 720P / 1080P · 2–15s · 30 fps · MP4 | Lower-cost drafts with an audio toggle |
wan2.6-r2v | Reference to video | Text + images or videos | 720P / 1080P · 2–10s · 30 fps · MP4 | Character appearance and voice continuity |
wan2.6-r2v-flash | Fast reference to video | Text + images or videos | 720P / 1080P · 2–10s · 30 fps · MP4 | Batch reference drafts at Flash pricing |
wan2.6-t2v-uswan2.6-i2v-us | US-scoped T2V / I2V | Text + optional audio | 720P / 1080P · 5 / 10 / 15s · 30 fps · MP4 | Data and inference restricted to the US scope |
wan2.6-imagewan2.6-t2i | Image generation and editing | Text + image | PNG · ≤ 1440 × 1440 · ≤ 4 images | Related still-image model; up to 4 outputs |
How the API works
Video jobs are asynchronous: submit once, poll the task ID, then save the returned asset before its temporary URL expires.
Create a task
Send the model, prompt, media and generation parameters to the regional synthesis endpoint.
Poll the result
Read the task instead of submitting duplicate jobs while it moves from pending to running.
Store the video
Download the MP4 after success and copy it to durable storage for production use.
POST /api/v1/services/aigc/
video-generation/video-synthesis
X-DashScope-Async: enable
GET /api/v1/tasks/{task_id}The API key, workspace endpoint and model must belong to the same region. Task IDs and result URLs remain available for 24 hours.
Estimate API cost
Use the calculator for Singapore international list prices. Standard T2V/I2V bill output only; reference endpoints also bill up to five seconds of input video.
Estimated list-price cost
Output seconds × resolution rate
USD · Singapore international deploymentList-price snapshot for August 28, 2026. Promotions can differ. Eligible Singapore models include a 50-second quota valid for 90 days under Model Studio's quota rules.
Wan 2.6 or Wan 2.7?
Wan 2.6 remains useful for Flash pricing and wider regional variants, while Wan 2.7 consolidates more production controls into its newer request format.
| Capability | Wan 2.6 | Wan 2.7 |
|---|---|---|
| Resolution parameters | size, plus shot_type for multi-shot | resolution + ratio; shots described in natural language |
| Image-to-video | First-frame generation | First frame, first/last frame and video continuation |
| Character voice | Voice can follow a reference video | A separate reference_voice can be assigned per subject |
| Video editing | No Wan 2.6 video-edit endpoint | wan2.7-videoedit for instruction editing and transfer |
| Cost option | Flash audio and silent modes | Standard $0.10 / $0.15 per second |
Continue exploring
Move between the model generations and the practical tools built around the Wan workflow.
Frequently asked questions
Is Wan 2.6 open source?
The model IDs on this page are hosted Alibaba Cloud Model Studio services. They require a regional API key and are separate from downloadable open-weight Wan releases.
Can Wan 2.6 generate sound?
Yes. T2V and standard I2V/R2V produce audiovisual video, while Flash endpoints let you choose audio or silent output.
Is there a free Wan 2.6 API?
It is not an anonymous free API. Eligible new Model Studio users can receive a 50-second quota for supported Singapore models, valid for 90 days under the official quota rules.
Which Wan 2.6 model should I choose?
Choose T2V for text-only scenes, I2V for a controlled first frame, R2V for reusable characters, and a Flash variant when lower cost and faster iteration are the priority.