Wan 3.0 video model
Wan 3.0 is an AI video model from Alibaba Cloud. In Hepifun it makes clips from 2 to 30 seconds at 480p, 720p or 1080p. Pricing is per second, in tokens.
What Wan 3.0 is
Wan 3.0 is a video generation model by Alibaba Cloud. In Hepifun you can start from text, from a photo (image to video), from first and last frames, or from reference images of your characters.
Hepifun is an independent service. We use Wan 3.0 through the official Alibaba Cloud Model Studio API.
Specs in Hepifun
- Resolutions: 480p, 720p, 1080p
- Clip length: 2 to 30 seconds
- Tiers: Standard and Prime
Price per second
| Tier | Resolution | Tokens per second | Price per second |
|---|---|---|---|
| Standard | 480p | 2 | Rp1,200 |
| Standard | 720p | 4 | Rp2,400 |
| Standard | 1080p | 8 | Rp4,800 |
| Prime | 480p | 3 | Rp1,800 |
| Prime | 720p | 6 | Rp3,600 |
| Prime | 1080p | 12 | Rp7,200 |
1 token = Rp600. See the full pricing page.
Example outputs
Real clips made through Hepifun on Wan 3.0 Standard. Each one shows its prompt, resolution, length and token cost.
“Cinematic aerial shot gliding slowly forward over terraced rice fields in Bali at sunrise, mist drifting between palm trees, golden light on the...”
“A fictional Indonesian street vendor in a red cap stands at a rainy night market stall and says cheerfully in English: "Hot corn, fresh and sweet!...”
“A short cinematic sequence of a fishing village at dawn, in three shots. Shot 1 [0-2s]: Wide shot of colorful wooden boats on calm water, mist at...”
How to use it
- Write a prompt, or upload a photo for image to video.
- Choose the resolution and length.
- Check the token estimate on the send button.
- Generate, then download. If the job fails, the tokens return in full.
Start with text to video or image to video.
Limits
- Jobs go into a queue. Each account can run a limited number of jobs at the same time, so extra jobs wait their turn.
- Maximum clip length is 30 seconds.
- Results are not always perfect. Hands, on-screen text and very fast motion can fail.
Wan 3.0 and Wan 2.7 in Hepifun
Wan 3.0 makes video. Wan 2.7, also by Alibaba Cloud, makes images in two options: Standard (2 tokens) and Pro (4 tokens).
Your questions
Can I use Wan 3.0 without paying? No. Every clip uses tokens. Top up first, then generate.
How long can a clip be? Each clip is 2 to 30 seconds.
Can I use my own photo? Only if you have the rights to it, and anyone recognizable in it has given permission. Read the Content Policy.
Does Wan 3.0 support audio? Yes. Wan 3.0 generates sound together with the video, such as ambience and speech, and you can write what should be heard in your prompt. Sound is included in the price. We have not transcribed the speech in our test clips, so how clearly spoken words come out can vary.