WAN 2.7: multi-reference AI video generation
WAN 2.7 comes from Alibaba's WAN AI video model family. It supports text-to-video, image-to-video, first-to-last frame control, up to 5 reference images, 4K resolution, and clips of 5, 10, or 15 seconds - all with roughly 30% cleaner output than its predecessor.
- Guide scenes with reference images
- Control first and last frames
- Create detailed video clips
What is WAN 2.7?
WAN 2.7 is an AI video generation model from Alibaba's WAN model family. Picsart made WAN 2.7 its default AI video model, running it across mobile native, mobile web, and desktop web. It supports every modality - text-to-video, image-to-video, first-to-last frame generation, up to 5 reference images, and audio synchronization — delivering noticeably cleaner output with sharper edges, more believable skin tones, and more grounded color balance.
WAN 2.7 capabilities
WAN 2.7 generates video at up to 4K resolution (4096×2160) in clips of 5, 10, or 15 seconds across all standard aspect ratios. It accepts up to 5 reference images as visual anchors - keeping characters, products, environments, and brand assets consistent throughout the video. Camera control responds to plain-language direction (pan, dolly, zoom), and audio sync supports ambient sound, dialogue-matched lip-sync, and background music in a single run.
What you can create with WAN 2.7
Generate multi-character video scenes
Create video scenes with multiple characters where every face and outfit stays consistent across the entire clip, powered by up to 5 reference images as visual anchors.
How Picsart uses WAN 2.7
Picsart runs WAN 2.7 as its default AI video model across mobile native, mobile web, and desktop web. Five reference images, first-to-last frame control, 4K output, natural-language camera direction, and tighter audio sync are all set as defaults across every video tool - no extra configuration needed.
WAN 2.7 is integrated into Picsart's AI Video Generator and AI Playground, where you can compare it with 130+ other AI models using a single prompt.
Why creators choose WAN 2.7
Creators choose WAN 2.7 for its multi-reference image support - up to 5 visual anchors that no other WAN model or competitor currently matches. Combined with 4K resolution, physically realistic motion, plain-language camera control, and integrated audio sync, it delivers the most versatile AI video generation workflow available in Picsart.
Understand video model choices
Learn how to compare video models and choose an output.
Explore more models like WAN 2.7
Compare WAN 2.7 with other video models for motion, ads, and social clips.
WAN 2.7 FAQ
WAN 2.7 is an AI video generation model from Alibaba's WAN model family. It supports text-to-video, image-to-video, first-to-last frame control, up to 5 reference images, 4K resolution (4096×2160), and clips of 5, 10, or 15 seconds with integrated audio synchronization.
Generate AI visuals with WAN 2.7
Pro
Most popularAI tools for everyday creative work.
Ultra
Most powerfulHeavy AI usage for creators & teams.