Alibaba
Newreleased August 2026Wan 3.0
Up to 30 s single-shot video with sound
Alibaba's latest Wan model generates up to 30 seconds in one pass from text or a starting image, in 480p to 1080p, with audio created together with the picture and optional prompt enhancement.
- 5–30 s clips in one pass
- Audio generated with the video
- 480p, 720p and 1080p
- Negative prompt and prompt enhancement
- Min. credits
- 50
- Typical time
- ~300s
- Max resolution
- 1080p
Specs
Capabilities
- Modes
- Text to videoImage to video
- Inputs
- TextPhoto
- Aspect ratios
- 16:99:161:14:33:4
- Durations
- 5s10s15s30s
- Resolutions
- 480p720p1080p
- Native audio
- Yes
Credits
Pricing
Credits per output. Free generations cover Standard quality.
| Setting | Quality | Credits |
|---|---|---|
| 480p · 5s | Draft | 50 |
| 480p · 10s | Standard | 100 |
| 720p · 5s | Standard | 100 |
| 480p · 15s | High | 150 |
| 1080p · 5s | High | 200 |
| 720p · 10s | High | 200 |
| 480p · 30s | High | 300 |
| 720p · 15s | High | 300 |
| 1080p · 10s | High | 400 |
| 1080p · 15s | Ultra | 590 |
| 720p · 30s | Ultra | 590 |
| 1080p · 30s | Ultra | 1180 |
