# Wan v2.6 Text-to-Video

The Wan 2.6 release advances Alibaba Cloud's text-to-video pipeline with longer clips, smarter scene composition, and sharper visual fidelity, making it the version to deploy when production-quality output is required.

- **Model ID:** `alibaba/wan-v2.6-t2v`
- **Type:** video
- **Providers:** alibaba
- **Context window:** Not applicable
- **Maximum output tokens:** 0
- **Pricing:** See the provider details below
- **Canonical page:** https://vercel.com/ai-gateway/models/wan-v2.6-t2v

## Supported parameters

- **Parameters:** `aspectRatio`, `resolution`, `duration`, `fps`
- **Operations:** `text-to-video`
- **Input modalities:** `text`, `audio`
- **Output modalities:** `video`
- **Aspect ratios:** `16:9`, `9:16`
- **Resolutions:** `720p`, `1080p`
- **Durations:** `5s`, `6s`, `7s`, `8s`, `9s`, `10s`, `11s`, `12s`, `13s`, `14s`, `15s`
- **Frame rates:** `24 fps`
- **Maximum outputs per request:** Not reported

## Example

```ts
import { experimental_generateVideo as generateVideo } from 'ai';

const result = await generateVideo({
  model: 'alibaba/wan-v2.6-t2v',
  prompt: 'A serene mountain lake at sunrise.'
});
```

## About

Part of Alibaba Cloud's Wan 2.6 update, this model is the production-grade text-to-video tier in that family. You provide a text prompt describing a scene, and the model returns a finished video clip, up to 15 seconds, at 720p or 1080p, in any of five aspect ratios covering landscape, portrait, square, and broadcast formats.

Intelligent multi-shot storytelling is the most distinct new capability. When a prompt describes a sequence of events spanning multiple locations or moments, the model introduces scene cuts and camera transitions on its own rather than forcing everything into a single continuous take. The visual identity of characters and objects stays consistent across these cuts, which means a product demo or short narrative can be generated as a cohesive piece rather than stitched together from separate clips in post-production.

On the rendering side, the 2.6 generation produces noticeably cleaner output than its predecessor. Frame-to-frame temporal consistency is improved, reducing the flicker artifacts that plagued earlier text-to-video models on detailed elements like text overlays, fine textures, and hair. Audio, ambient, effects, and music, continues to be generated in the same pass as the video, maintaining the integrated pipeline introduced in the 2.5 generation.

## What to consider

Running a few test prompts in the AI Gateway playground before full integration will help you calibrate resolution-based pricing and generation times across available providers.

## When to use

### Best For

- Production-quality short-form video for advertising, social content, or product demos where visual polish matters
- Narrative prompts that span multiple scenes, the automatic multi-shot composition removes the need for manual clip editing
- Broadcast or traditional media workflows using the 4:3 and 3:4 aspect ratios not available in the 2.5 preview
- Any text-to-video task where 15-second duration and 1080p resolution represent your quality target

### Consider Alternatives When

- If you want to animate an existing image rather than generate from text alone, wan-v2.6-i2v is designed for that workflow
- If character consistency must be locked to a reference video of a real person, wan-v2.6-r2v handles reference-based synthesis
- If your budget is tight and 480p output is acceptable for your use case, wan-v2.5-t2v-preview supports the lower resolution tier

## Best for

- **Production-quality short-form video:** Advertising, social content, and product demos where visual polish matters
- **Multi-scene narrative prompts:** Automatic multi-shot composition removes the need for manual clip editing
- **Broadcast and traditional media:** Workflows using the 4:3 and 3:4 aspect ratios not available in the 2.5 preview
- **15-second 1080p targets:** Any text-to-video task where this duration and resolution represent your quality ceiling

## Consider alternatives

- **Animating an existing image:** wan-v2.6-i2v is designed for image-driven workflows rather than pure text-to-video
- **Real-person reference consistency:** wan-v2.6-r2v handles reference-based synthesis when character identity must match a source video
- **Tight budget at 480p:** wan-v2.5-t2v-preview supports the lower resolution tier when 480p output is acceptable

## Frequently asked questions

### What makes multi-shot storytelling different from generating multiple separate clips?

The model handles scene transitions internally, maintaining character and object consistency across cuts. This produces a single cohesive video rather than requiring you to generate and stitch together separate clips with potential visual discontinuities.

### Which aspect ratios are new in the 2.6 release?

The 2.6 model adds 4:3 and 3:4 to the existing 16:9, 9:16, and 1:1 options, covering traditional broadcast and print-adjacent formats that the 2.5 preview did not support.

### How does the temporal stability improvement affect output quality?

The 2.6 generation reduces frame-to-frame flicker on high-frequency visual details, text overlays, logos, hair, and fine textures render more cleanly than in the 2.5 preview, producing more professional-looking output.

### Can I specify the number or timing of scene cuts in a multi-shot video?

The model determines scene structure automatically based on the prompt content. You influence the number of scenes through the narrative structure of your prompt rather than through an explicit parameter.

### What duration options are available?

Clips can be generated at 5, 10, or 15 seconds. The 15-second option is the longest in the Wan 2.6 series.

### Is 480p output available with this model?

No. Wan v2.6 Text-to-Video supports 720p and 1080p only. For 480p draft-quality work or cost savings, use the Wan v2.5 T2V Preview.

## Links

- [Model page](https://vercel.com/ai-gateway/models/wan-v2.6-t2v)
- [AI Gateway documentation](https://vercel.com/docs/ai-gateway)
- [Provider model documentation](https://wan.video/)
- [Provider pricing](https://modelstudio.console.alibabacloud.com/ap-southeast-1/?tab=doc#/doc/?type=model&url=prices)


If you encounter incorrect information, can’t find what you need, need a capability Vercel doesn’t currently offer, or have any other feedback to report, please do it by following the instructions at https://vercel.com/feedback.md.