Wan 3.0 API
Wan 3.0, Alibaba's newest video model, via WaveSpeed — 720p clips of 2–30 seconds with generated audio, text-to-video and image-to-video, at $0.10/s. Billed in Varosity Credits.
$0.10/s on Varosity credits. Clips up to 30s with native audio.
no wavespeedaccount needed · provider's list price, no markup · failed renders never billed
Best for
- newest Wan, with audio
- single takes of 2 to 30 seconds
- text- or image-to-video
Trade-offs
- 720p via Varosity
How to get the best result
Wan 3.0, Alibaba's newest video model: 720p clips with generated audio and single takes of 2 to 30 seconds, from text or a first-frame image. Pick instead: wan-2.6 for the previous Wan on fal; kling-3.0 for fast 5 or 10s cinematic clips; veo-3.1 for 4-8s lip-synced talking heads.
Prompting
- Describe the subject, the action, the setting, then the camera movement and lighting.
- Describe the sound you want (ambience, music, a spoken line in quotes); audio is generated with the picture.
- Image-to-video: describe the motion, not the picture; leave aspectRatio unset to follow the image.
Settings on Varosity
- durationSec is whole seconds from 2 to 30, billed per second produced at $0.10/s.
- Always 720p; aspectRatio must be 16:9, 9:16 or 1:1 (other values are refused before anything is charged).
- With referenceImageUrl the image is the first frame and its framing is kept.
- A 2-second clip took about 70 seconds in testing (2026-10-25); longer clips take longer, so keep polling get_job.
Avoid
- Asking for 30 seconds when 5 will do: cost and render time scale with every second.
- Relying on readable text in the scene.
Example prompt
A fisherman hauls a glowing net onto a wooden boat at dawn, slow push-in, mist on the water, gulls crying, waves lapping the hull.
From wavespeed.ai, wavespeed.ai, checked against what Varosity sends.
Call it from your code
One Varosity key works for every model. Video renders are async — the response has a jobId; poll GET /api/v1/jobs/{jobId} until status is "succeeded", then read outputUrl.
curl -X POST https://varosity.ai/api/v1/video/generate \
-H "Authorization: Bearer $VAROSITY_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "cinematic aerial of a coastal road at golden hour, slow push-in",
"modelId": "wan-3.0",
"aspectRatio": "16:9",
"durationSec": 5
}'Or ask your agent
Connect Claude, Cursor, or any MCP client to https://varosity.ai/api/mcp with your key, then say:
"Use Varosity with wan-3.0 to make a 5-second clip of …"Agent setup guide
Alternatives to Wan 3.0
- Kling 3.0 Promulti-shot consistency · cinematic motion$0.168/s
- Kling 4.0newest Kling, native audio · single takes up to 30s$0.157/s
- MiniMax H3MiniMax's newest video model, with sound · 4 to 15 second takes$0.10/s
- Seedance 2.0top-ranked motion + physics · cinematic camera control$0.3024/s
- Seedance 2.0 Fastnear-flagship quality at lower cost · faster renders$0.2419/s
- Veo 3.1 (Quality)state-of-the-art fidelity · best lip sync$0.40/s