Wan 3.0 AI video generator
Make video with Wan 3.0 Prime, the video model from Alibaba: clips of up to 30 seconds at up to 1080p, with sound rendered in, from a written prompt or a photo. Wan 3.0 Prime First-Last Frame animates between two photos you choose.
- Up to 30 second clips
- Up to 1080p
- Sound included
- From 35 credits a clip
Made with Wan on QynTro
Nothing here is a stock sample: each one was made on QynTro, and its page shows the exact prompt, so you can make it again or change it.
SHOT 1 of 2, 10 seconds: Night, heavy rain, a stone temple courtyard lit by warm paper lanterns. The young woman warrior stands alone in the centre while six masked warriors in…
Full prompt and settings →The instruction on this one is the pair of frames below. The first is a woman standing still in waist-high summer grass with the sun going down through the treeline behind her…
Full prompt and settings →Image 1 is the frame this opens on. The woman walks slowly on through the tall grass, drawing her open hand along the tops of it; where her fingers pass, the seed heads burst and…
Full prompt and settings →Image 1 is the frame this opens on. The little girl is mid-spin on the sunlit living-room floor; let the turn carry round and come gently to rest — her arms lower, her lifted foot…
Full prompt and settings →Image 1 is the frame this opens on. The girl is resting her chin on her folded arms at the rain-streaked window; she breathes out once onto the cold glass, a small patch of fog…
Full prompt and settings →Image 1 is the frame this opens on. The knitted girl is mid-twirl in the felt meadow; let the turn carry round and settle — her woollen plaits swing wide and drop against her…
Full prompt and settings →Image 1 is the frame this opens on. The woman raises the brass lantern a little higher and takes one slow step toward the dark trees, turning her head to look into them. Keep her…
Full prompt and settings →Image 1 is the frame this opens on. The boy and the terrier are both in the air mid-jump in the farmhouse kitchen; let them come down together — his bare feet find the terracotta…
Full prompt and settings →Which Wan to pick
Wan charges by the second and by resolution: 720p costs twice 480p and 1080p four times, so the table shows the 5 second 480p price. Sound is always included.
| Model | Best for | Text to video | Image to video | Length | From | Try it |
|---|---|---|---|---|---|---|
| Wan 3.0 PrimeNEW | The newest Wan: up to 30 seconds at 1080p, with sound. | Yes | Yes | 5 to 30 s | 35 credits · 5 s clip | Try → |
| Wan 3.0 Prime First-Last FrameNEW | Give it a first and a last photo, and it makes the motion between them. | No | Yes | 5 to 30 s | 35 credits · 5 s clip | Try → |
Prices are in media credits and are the ones the tool charges today; the tool shows the exact number before you make anything. A failed generation gives its credits back. See plans and credit packs
From an idea to a Wan clip in three steps
- 1
Describe the scene
Write what happens and how the camera moves. Or start from a photo, or from a first and a last photo, and describe the motion.
- 2
Pick a length and resolution
Choose 5, 10, 15, 20 or 30 seconds and 480p, 720p or 1080p. The tool shows the exact credits before you start.
- 3
Generate and download
The clip lands in your library with its sound, ready to download. If a render fails, its credits go straight back.
What people make with Wan
Longer stories
Up to 30 seconds in one clip: room for a whole moment, not just a shot.
Scenes with sound
Wan renders the audio with the picture, so a clip arrives ready to post.
Before and after
Animate from a first photo to a last one with First-Last Frame.
Cheap drafts
Try an idea at 480p, then make the keeper at 1080p.
Wan questions
What is Wan?
How long can a Wan video be?
Does Wan make sound?
Can Wan animate between two photos?
Can I try Wan for free?
How does Wan compare with Kling?
Do I need an Alibaba account?
Other models on QynTro
AI video model · ByteDance
Seedance
ByteDance video with sound built in, from a prompt or a photo.
Open Seedance →
AI video model · Kuaishou
Kling
Cinematic text-to-video and image-to-video: Kling 3 Pro and O3.
Open Kling →
AI video model · MiniMax
MiniMax
H3 Max video up to 15 seconds and 1080p, with camera moves.
Open MiniMax →
AI video model · Google
Gemini Omni Flash
Google's fast video model: clips with sound, up to 15 seconds and 4K.
Open Gemini Omni Flash →
Make your first Wan clip
One QynTro account opens every model on this page, plus the rest of the image, video, music and voice tools.