Alibaba
WAN 3.0 Prime Reference: AI video generation
The higher quality tier of WAN 3.0 driven by reference images, videos and audio, up to 30 seconds at 480p to 1080p with native audio.
Free credits on signup, no card required. Plans from $19 one time.

Example videos made on Enhance AI
These videos come from related models on the same account. The model that made each one is named under it.
A tabby cat wearing a small blue scarf walking along a garden wall, then jumping down onto a lawn. Low angle tracking shot, golden hour light, warm colour grade. Cinematic and photorealistic, natural motion. Photographic, natural, no text, no lettering, no captions, no subtitles, no watermark, no logo, no brand names, no product labels, no user interface, no screens, no recognisable fictional characters, no celebrities or recognisable real people.
Generated with MiniMax H3 Max
A woman with a bright orange backpack hiking a ridge trail, turning to look at the valley. Static wide shot on a tripod, harsh midday sun, deep shadows. Cinematic and photorealistic, natural motion. Photographic, natural, no text, no lettering, no captions, no subtitles, no watermark, no logo, no brand names, no product labels, no user interface, no screens, no recognisable fictional characters, no celebrities or recognisable real people.
Generated with MiniMax H3 Max
A boy in a yellow raincoat splashing through puddles on a cobbled lane. Slow pull back revealing the wider scene, crisp clear daylight, vivid colours. Cinematic and photorealistic, natural motion. Photographic, natural, no text, no lettering, no captions, no subtitles, no watermark, no logo, no brand names, no product labels, no user interface, no screens, no recognisable fictional characters, no celebrities or recognisable real people.
Generated with MiniMax H3 Max
A white sailboat with a striped sail rounding a buoy in a bay. Smooth tracking shot moving alongside the subject, moody low key lighting, fine film grain. Cinematic and photorealistic, natural motion. Photographic, natural, no text, no lettering, no captions, no subtitles, no watermark, no logo, no brand names, no product labels, no user interface, no screens, no recognisable fictional characters, no celebrities or recognisable real people.
Generated with MiniMax H3 Max
A golden retriever carrying a red ball along a pier, tail wagging. Slow orbit around the subject, blue hour, cool tones with warm highlights. Cinematic and photorealistic, natural motion. Photographic, natural, no text, no lettering, no captions, no subtitles, no watermark, no logo, no brand names, no product labels, no user interface, no screens, no recognisable fictional characters, no celebrities or recognisable real people.
Generated with MiniMax H3 Max
A man in a grey wool coat and round glasses reading on a park bench as leaves fall. Aerial view gliding forward, misty dawn light, desaturated greens. Cinematic and photorealistic, natural motion. Photographic, natural, no text, no lettering, no captions, no subtitles, no watermark, no logo, no brand names, no product labels, no user interface, no screens, no recognisable fictional characters, no celebrities or recognisable real people.
Generated with MiniMax H3 Max
Key features
WAN 3.0 Prime Reference turns reference images into video inside Enhance AI. It produces clips from 5 to 30 seconds at a resolution you choose, 720p by default. On top of that it renders 16:9, 9:16, 1:1, 4:3, 3:4 and adaptive and it accepts up to 10 reference images. WAN 3.0 Prime Reference is built by Alibaba. On Enhance AI it runs on the same account as every other model, with no separate subscription.
Clip length
WAN 3.0 Prime Reference renders clips from 5 to 30 seconds, and you pick the length before you render. Longer clips cost more because video is billed by length, so rough out an idea short and only render the full length once the shot works.
Resolution
Choose 480p, 720p or 1080p before you render. A lower resolution costs less, so test a prompt low and rerender the keeper at the top setting. The studio shows the price for the setting you picked before anything is charged.
Input modes
WAN 3.0 Prime Reference takes a written prompt, reference images, reference videos (optional) and reference audio (optional). Add them in the studio, check the preview and render. Every output lands in your history, ready to download or to pass to another model on the same account.
Reference images
Add up to 10 reference images and WAN 3.0 Prime Reference holds a character, a product or a style steady across shots: the gap between a clip that looks like your brand and one that looks like the model's idea of it. Attach them next to the prompt.
Aspect ratios
WAN 3.0 Prime Reference renders 16:9, 9:16, 1:1, 4:3, 3:4 and adaptive. Pick the ratio for the platform before you render rather than cropping afterwards: 9:16 for Reels and Shorts, 16:9 for YouTube and 1:1 for feeds. The whole frame is generated for that shape.
Audio
WAN 3.0 Prime Reference can generate sound with the picture, so the clip arrives with audio that matches what is on screen rather than silence you have to fill later. You can still mute it and lay your own voice or music track over the top in any editor.
Specifications
- Provider
- Alibaba
- Type
- Video
- Released
- 24 August 2026
- Resolution
- 720p by default
- Duration
- 5 to 30 seconds
- Aspect ratio
- 16:9, 9:16, 1:1, 4:3, 3:4, adaptive
- Reference images
- Up to 10
How to use WAN 3.0 Prime Reference on Enhance AI
- 1
Sign in and pick WAN 3.0 Prime Reference
Open the Video Studio on Enhance AI and choose WAN 3.0 Prime Reference from the model list. A new account starts with free credits and no card is needed. The Try button on this page opens the studio with WAN 3.0 Prime Reference already selected.
- 2
Add references and describe the shot
Upload your reference images (up to 10) and describe the shot they should appear in: the setting, the action and the camera move. Then pick a length from 5 to 30 seconds, a resolution and an aspect ratio.
- 3
Generate, then download or extend
Press Generate. WAN 3.0 Prime Reference renders in the cloud and the clip lands in your history on Enhance AI, where you can download the MP4, upscale it or extend it with another model on the same account. You see the price before you confirm.
What people make with WAN 3.0 Prime Reference
A consistent character
The hardest thing in generated video is the same face in every shot. Give WAN 3.0 Prime Reference reference images of the character and it holds the look across a whole scene: the same features, the same costume, the same proportions. That is what turns a set of clips into a story.
A product across scenes
Upload photos of the product and describe each setting, and WAN 3.0 Prime Reference places the same bottle, shoe or device in every shot with its shape and label intact. Build a full campaign of scenes from a single set of studio photos rather than shooting each location.
Style from a mood board
References do not have to be a subject. Feed WAN 3.0 Prime Reference frames from a film, a set of photographs or your own past work, and the model carries the palette, the grain and the framing into new shots. It is the fastest way to match a brand look or a client's reference reel.
Common questions
- What is WAN 3.0 Prime Reference?
- WAN 3.0 Prime Reference is a reference to video model from Alibaba. It builds a new shot from up to 10 reference images and a written prompt. On Enhance AI it runs on the same account as 250+ other models, with no separate subscription.
- What do I need to upload to use WAN 3.0 Prime Reference?
- A written prompt, reference images, reference videos (optional) and reference audio (optional). Add them in the studio on Enhance AI, check the preview and render. The output lands in your history, ready to download or to pass to another model on the same account.
- When was WAN 3.0 Prime Reference released?
- Alibaba released WAN 3.0 Prime Reference on 24 August 2026. You can run it on Enhance AI now, along with every other video model on one account.
- What is new in WAN 3.0 Prime Reference compared to WAN 2.7 Reference?
- On Enhance AI, WAN 2.7 Reference renders clips up to 10 seconds and takes up to 4 reference images. WAN 3.0 Prime Reference renders clips up to 30 seconds, takes up to 10 reference images and adds native audio. Both run on the same account, so you can pick whichever fits the shot and the budget.
- How long can a WAN 3.0 Prime Reference clip be?
- WAN 3.0 Prime Reference produces clips from 5 to 30 seconds. Longer runs cost more, because video is billed by length rather than by generation.
- What resolution does WAN 3.0 Prime Reference output?
- WAN 3.0 Prime Reference renders at a resolution you choose, 720p by default. You pick the resolution before you render, and a lower one costs less.
- Does WAN 3.0 Prime Reference take reference images?
- Yes, up to 10. Use them to hold a character, a product or a style steady across shots.
- Does WAN 3.0 Prime Reference generate audio?
- Yes. WAN 3.0 Prime Reference can generate sound with the picture, so the clip arrives with audio rather than silence. Mute it and lay your own track over the top if you prefer.
- Is WAN 3.0 Prime Reference free to try?
- Yes. Free credits on signup, no card required, so you can run WAN 3.0 Prime Reference before paying anything. Plans from $19 one time when you need more, and one account covers WAN 3.0 Prime Reference and 250+ other models.
- What does it cost to run WAN 3.0 Prime Reference?
- Video renders are billed from your wallet by clip length and resolution, so a short clip at a lower resolution costs less than a long one at full quality. You see the price before you render.
- Do I need a separate subscription for WAN 3.0 Prime Reference?
- No. WAN 3.0 Prime Reference runs on the same Enhance AI account as every other model, so you are not paying one provider for video and another for images.
More from Alibaba
- WAN 2.7 ReferenceWAN 2.7 reference to video generation.View
- WAN 2.6 ReferenceWAN 2.6 reference to video generation.View
- WAN 3.0Alibaba's newest video model. Up to 30 seconds with native audio, from 480p to 1080p.View
- WAN 3.0 PrimeThe higher quality tier of WAN 3.0. Same controls, sharper results: up to 30 seconds with native audio, from 480p to 1080p.View
- WAN 3.0 VibeAlibaba's newest video model with a deep thinking mode for complex prompts. Up to 30 seconds with native audio, from 480p to 1080p.View
- WAN 3.0 Vibe ProThe higher quality tier of WAN 3.0 Vibe, with a deep thinking mode for complex prompts. Up to 30 seconds with native audio, from 480p to 1080p.View
Other reference to video models
More from Enhance AI