Unlimited Seedance 2.5 is live on Imagine with 50% OFF on Plans
Wan 2.7 AI Video Generator
Wan 2.7 AI video generator creates cinematic, realistic and stylized animated videos with text and image inputs for up to 15 seconds durations. It offers 720p and 1080p resolutions. Use this innovative 9-image grid system to create cinematic series, short films, product shoots and advertisements.
How to use Wan 2.7 AI Video Generator?
Enter a prompt
Write your prompt or upload a reference image. Feed up to 5 images to guide the output. Be descriptive about setting, object, actions, scene, mood, aesthetics and style.
Choose Your Output Settings
Choose your desired resolution: 720p for fast output, 1080p for high-quality results. Set video duration for up to 15 seconds. Select from 5 aspect ratios: 16:9, 9:16, 1:1, 4:3, or 3:4 based on where you're publishing.
Generate And Download
Click generate. Wan 2.7 AI video generator will create video with strong prompt accuracy. Download the video, post to platforms or share with ImagineArt Community.
Strong T2V and I2V Workflow
Whether you're starting from a blank page or an existing photo, Wan 2.7 handles both paths with equal precision. Describe a scene from scratch, or animate a reference image directly, Wan 2.7's image to video pipeline turns a still photo into motion without losing the original composition. Creators and brands can generate product demos, social content, and narrative scenes from prompt or photo to polished video in a single generation, without switching tools between the concept and the final render.
Audio-Video Synced Generations
Sound and motion come out of the same generation, not a separate pass you have to align afterward. Wan 2.7 generates synchronized audio alongside the visuals, no post-production alignment needed. The model drives lip movements and scene pacing directly from the prompt and sound, so talking head videos, dialogue scenes, and voice-driven brand content arrive already in sync. For projects that need additional voice work or a custom soundtrack layered in afterward, the Audio Studio picks up from there.
Cinematic Motion Quality
Motion is where a generated video either looks real or falls apart, and it's a specific strength of this model. Wan 2.7 produces smoother, more physically plausible motion, generating body movement with natural weight, directed camera behavior, and character detail that holds across skin, fabric, and hair. Cinematic pitches, lifestyle content, and high-fidelity ads come out sharp on the first pass, without needing a second or third generation to fix warped movement. Once you're happy with the raw output, the AI Video Editor is where to fine-tune pacing, add transitions, or trim the final cut.
9-Grid Image-based Storyboarding
Wan 2.7's most distinctive feature is not just a resolution or speed improvement, it's a genuinely different way to guide a generation. The 9-grid system accepts up to nine reference images in a structured 3×3 layout, multi-angle shots, sequential poses, scene variants. Wan 2.7 uses all of them together to lock subject identity, pose accuracy, and composition across every frame of the generated video. This makes it a strong fit for product shoots that need consistent framing across several angles, or a short cinematic series that needs the same character to hold up shot to shot. For a broader character-consistency comparison across ImagineArt's models, see AI video character consistency.
Trusted by Professionals and Creators from top Brands and Companies
FAQs
Wan 2.7 is Alibaba's latest open-source AI video model for creators who need production-ready results without a production budget. It supports text-to-video, image-to-video, multi-reference generation, first-to-last frame control, instruction-based editing, and video recreation at up to 1080p.
On ImagineArt, Wan 2.7 generates videos at 720p or 1080p, with duration options of 5, 10, or 15 seconds. Five aspect ratios are available: 16:9, 9:16, 1:1, 4:3, and 3:4.
Wan 2.7 is available on ImagineArt for global users. Go to imagine.art/apps/wan-2.7-ai-video, sign in or create a free account, and start generating immediately.
Yes, Wan 2.7 is released as an open-source model by Alibaba. On ImagineArt, you can use it directly through the hosted interface without needing to set up the model yourself.
Wan 2.7 adds the 9-grid image-based storyboarding system and improved audio-video sync over Wan 2.6.