Wan 3.0 — 30-Second AI Video with Native Audio
Alibaba's newest Wan model on Imagenter AI. Generate up to 30 seconds in a single shot at up to 1080p, with sound generated alongside the picture. Start from text, a first and last frame, or your own images, clips and audio. Free trial credits on signup.
What Wan 3.0 Does Well
Two ways to take control of the shot:
First and Last Frame Control
Multimodal Reference
What Is Wan 3.0?
Wan 3.0 is the third generation of Alibaba's Wan video model, in public beta since August 2026. It takes text, images, video and audio as input and returns up to 30 seconds of video with synchronized native audio. On Imagenter AI it runs in three modes behind one page.
- 30 Seconds in One ShotAny length from 2 to 30 seconds in a single generation, with no stitching between clips.
- Native AudioScore, dialogue and ambience are generated with the picture and timed to the action.
- Normal or PrimePrime is a faster channel at a higher rate. The options and output quality are the same.
How to Use Wan 3.0
From idea to finished clip in four steps:
Wan 3.0 Features at a Glance
Everything the Wan 3.0 generator supports on Imagenter AI:
2 to 30 Second Clips
Any whole-second length in that range, at 480p, 720p or 1080p.
Native Audio
Soundtrack, dialogue and ambience generated in the same pass as the video.
First and Last Frame
Set where the shot starts and ends; Wan 3.0 fills in the motion between.
10 Images, 5 Videos, 5 Audio Clips
Reference inputs per generation, each given a role in the prompt.
Five Aspect Ratios
16:9, 9:16, 1:1, 4:3 or 3:4. Image to Video can follow your first frame instead.
Pay Per Second
From 15 credits per second, the estimate is shown before you generate, and failed generations are refunded.
FAQs about Wan 3.0
Common questions about running Wan 3.0 on Imagenter AI
What is Wan 3.0?
Wan 3.0 is Alibaba's newest AI video model, in public beta since August 2026. It generates up to 30 seconds of video with native audio from text, images, video and audio.
How much does Wan 3.0 cost?
On Normal speed, 15, 30 or 60 credits per second at 480p, 720p or 1080p, so a 5-second 720p clip costs 150 credits. Prime costs 25% more in Text and Image modes and 50% more in Reference mode. In Reference mode, the length of your reference videos is added to the billed seconds. The exact cost is shown before you generate, and new accounts get free credits to try it.
Is Wan 3.0 free?
Signing up for Imagenter AI is free, and new accounts get free trial credits, enough for one 2-second 480p Wan 3.0 clip on Normal speed (30 credits). A 5-second 480p clip costs 75 credits, and credit packs start at $10 for 1,000 credits. Credits never expire, and failed generations are refunded.
Wan 3.0 vs Wan 2.7 or Wan 2.5: which should I use?
Choose Wan 3.0 for clips longer than 15 seconds or for generating from several reference images, videos and audio clips: it runs 2 to 30 seconds at 15, 30 or 60 credits per second at 480p, 720p or 1080p on Normal speed. Wan 2.7 (30 or 45 credits per second at 720p or 1080p, up to 15 seconds) adds Video Edit and Video Extend for footage you already have. Wan 2.5 is the cheapest per second, about 13, 27 or 40 credits at 480p, 720p or 1080p, for clips up to 10 seconds.
How long can a Wan 3.0 video be?
2 to 30 seconds in a single generation. In Reference mode, reference videos can total up to 15 seconds, and their length plus the output length must stay within 30 seconds.
What's the difference between Normal and Prime?
Prime runs on a faster channel, so results come back sooner. The settings and output quality are the same; you pay extra for turnaround, not for better video.
How does Reference to Video work?
Upload up to 10 images, 5 videos and 5 audio clips, then cite them in the prompt as Image 1, Video 1 or Audio 1 and assign each a role, for example "Image 1 sits on the chair in Image 2".
Is Wan 2.6 still available?
Wan 2.6 has been retired on Imagenter AI. Wan 3.0 is the newer generation, with clips up to 30 seconds in a single shot. Wan 2.7 and Wan 2.5 are still available.
