Wan 2.6 — Multi-Shot AI Video with Native Audio
Wan 2.6, released by Alibaba's Wan team in December 2025, generates up to 15 seconds of video with native audio and can split a prompt into several shots. It can also cast a character from your reference images or videos into a new scene.
- Alibaba Wan 2.6
- Up to 15 s with native audio
- Multi-shot storytelling
- Reference-to-video (R2V)
More models
Why Wan 2.6 Is a Great AI Video Generator
Wan 2.6 was one of the first models to make 15-second, multi-shot clips with sound feel routine. It is a good choice for short narrative scenes and for putting a recurring character into new settings.
Multi-shot from one prompt
Describe a sequence and Wan 2.6 cuts it into shots that flow together.
15 seconds with sound
Longer clips with dialogue and ambience generated natively.
Reference-to-video
Cast a person or object from up to 5 reference images or clips into a new video.
Text, image or reference
Three ways to start, all in one model.
What Wan 2.6 can do
Text to video
Write the scene, the shots and the sound you want.
Image to video
Animate a still as the opening frame of the clip.
Reference to video
Mix reference images and short videos to keep a subject consistent.
5–15-second clips
Pick a length from 5 to 15 seconds, with sound generated in the same pass.
720p and 1080p output
Choose 720p and 1080p, in 16:9 format.
Reference inputs
Besides a text prompt you can add a first frame and up to 5 reference files.
Use cases
Starting points you can copy into the generator above.
| Use case | Good for | Example prompt |
|---|---|---|
| Short dramas | A scene with a few cuts and dialogue. | Shot 1: a woman opens a letter at a kitchen table. Shot 2: close-up of her eyes filling with tears. Shot 3: she smiles and says "He is coming home." Quiet piano |
| Recurring mascots | Your character in a new setting. | The character from the reference images skateboards through a sunny park, waves at the camera, cheerful music |
| Explainers | Short narrated scenes. | A teacher at a chalkboard draws the water cycle while explaining "Water evaporates, forms clouds, then falls as rain." Classroom ambience |
| Social stories | Vertical clips for feeds. | Vertical: a street food vendor flips noodles over a roaring flame at night, sizzling sounds and crowd chatter |
Wan 2.6 in 3 steps
- 1
Describe the shot or add a frame
Write what happens, how the camera moves and how it should sound. Add a start image or references to lock the look.
- 2
Pick length, resolution and ratio
Set duration, resolution and aspect ratio. The credit cost is shown before you submit.
- 3
Generate and keep your results
Videos appear in the generator and are saved to My works. Anything that fails is refunded automatically.
X Posts About Wan 2.6 — Multi-Shot AI Video with Native Audio
- Alamin@iam_chonchol
Wan 2.6 just raised the bar. Multi-shot storytelling, longer 15s flow, and clean video references — this is how AI video starts feeling cinematic, not stitched together. https://t.co/P04uLe8YvO
View on X - 1LittleCoder💻@1littlecoder
🔥 China is cooking - 15 seconds with Native Audio! Alibaba's Wan 2.6 is here! ✅Video duration from 3 to 15 seconds ✅Video resolution of 480p, 720p, or 1080p ✅ intelligent prompt rewriting https://t.co/n37ijoKrH8 pic.twitter.com/yK4PxZRM88
View on X - Wan@Alibaba_Wan
Wan2.6: Be What You Wanna Be Wan2.6 R2V is now available. 🎬️Supports recording in real time or uploading a 5 second reference video, and replicating the person, animal, animated character, or object from that clip into new videos. 🎬️Supports both single- and multi-character… pic.twitter.com/8ljfTQEsyk
View on X
YouTube Videos About Wan 2.6 — Multi-Shot AI Video with Native Audio
🎬 WAN2.6 — Generate multi-shot, native storyboards with one click!
Alibaba Cloud · YouTubeWan 2.6 Live Release | EU Session: The Cinematic Multimodal Generation Model
Tongyi Lab · YouTubeWan 2.6 Filmmaking: 15 Second Scenes From ONE Prompt (Multi Shot + Dialogue)
AI Creator Tools (ACT) · YouTubeWan 2.6 — Multi-Shot AI Video That Actually Looks Real (Full Breakdown)
Tech with Tas · YouTubeWAN 2.6 Showcase – Text-to-Video, Image-to-Video & V2V Tested
Eye For AI · YouTubeLonger AI Video + Multiple Shots + Built-in Dialogue
Excelerator · YouTube
Frequently asked questions
Longer 15-second clips, native audio, multi-shot storytelling from one prompt, and reference-to-video that casts subjects from your images or videos.
Up to 5 reference images or videos in reference-to-video mode; image-to-video uses one first frame.
From 94 to 158 credits per video, depending on duration, resolution. The exact cost is shown before you submit, and failed generations are refunded automatically.
720p and 1080p; aspect ratio 16:9. Clips run 5–15 seconds.
No. Wan 2.6 is developed by Alibaba Tongyi Lab; Nano Banana is an independent service that runs it through a model provider.
Ready to Animate with Wan 2.6?
Text, first-frame and reference-to-video in 1080p, clips up to 15 seconds with audio.
- Clips up to 15s
- Native audio
- Failed outputs refunded
