Kling v3.0
Native audio-supported video generation powered by Kling’s 3.x motion engine.
Now officially live and available to all public community creators and usersMarch 2026
Wan 2.7 AI Video Generator
Alibaba’s latest video generation model, Wan 2.7 delivers first/last frame control, multi-reference input support, and instruction-driven editing. Create 5 to 15-second clips at 720P or 1080P resolution.
Content policy notice
Any non-compliant generation will fail — this is not a scam or a site error. The I2V model has recently tightened content restrictions, and all real human face inputs (selfies, portraits, celebrities) are now blocked at the model level. Violent and NSFW content is also rejected. You can still create amazing videos with illustrations, anime characters, AI-generated faces, or stylized artwork.
Key features of Wan 2.7
What sets Wan 2.7 apart
Wan 2.7 brings first/last frame control, multi-reference input support, instruction-driven editing, and 15-second maximum clip lengths to Alibaba’s existing video generation lineup.
First / Last Frame Control
Lock in your desired opening and closing shots. Wan 2.7 automatically generates the seamless motion between those two points, giving you precise cinematic control without needing to master complex prompt workflows.
Lock in both your starting and ending compositions before generation starts
Perfect for product reveals, character transitions, and clean scene cuts
Eliminates the guesswork of hitting a specific final shot or pose
Multi-Reference Input
Upload up to 5 reference videos at once, and let the model use them to guide character appearance, environment details, and overall motion style for your output.
Upload up to 5 reference videos to direct your generated content
Preserve consistent visual styling for characters and environments throughout your clip
Ideal for brand campaigns, fashion showcases, and product-focused content that needs continuity
Instruction-Based Editing
Modify existing video clips using plain natural language. Tweak backgrounds, adjust lighting, change clothing, or shift style entirely without having to rebuild the clip from scratch.
Outline your desired changes in simple text, no complex timeline editing needed
Swap backgrounds, update outfits, or adjust lighting using a prompt
Quickly iterate on your clips without losing the original core motion
Up to 15 Seconds
Create clips up to 15 seconds in length, three times longer than previous Wan model iterations. Perfect for full product demos or short, cohesive scenes.
Choose clip lengths of 5, 10, or 15 seconds based on your project needs
Supports 480P, 720P, and 1080P output resolutions
Compatible with both 16:9 landscape and 9:16 portrait aspect ratios
Explore more models
Other Models to Try
Compare Wan 2.7 against other leading video generation models available on this platform.
Kling v3.0 Pro
Professional-level Kling 3.x output with enhanced fidelity and fine-tuned detail.
Hailuo 02
MiniMax’s newest video generation model, featuring robust motion dynamics and realistic output.
PixVerse V6: Professional AI Video Generator
PixVerse V6 stands as PixVerse’s most advanced generation model yet, boasting 20+ cinema-grade camera controls, built-in audio synchronization, a multi-shot sequencing engine, and support for 15-second 1080p clips. Create cinematic videos directly from text prompts or reference images.
FAQs
FAQ
All About I2V and Our Official Platform
What is Wan 2.7?
Released by Alibaba Tongyi Lab in March 2026, Wan 2.7 is their newest video generation model. Building on Wan 2.6, it adds first/last frame control, support for up to 5 concurrent reference video inputs, 9-grid image inputs, instruction-driven editing, and more accurate motion physics.
What is first/last frame control in Wan 2.7?
First/last frame control, often shortened to FLF2V, allows you to define both the opening and closing shots of your video. Wan 2.7 automatically fills in the smooth motion between those two fixed points, putting precise cinematic control at your fingertips: pick a starting composition and an ending scene, and the model handles the seamless interpolation of action.
How long can videos be with Wan 2.7?
Each clip created with Wan 2.7 can run between 2 and 15 seconds total — a major jump from prior Wan models that limited outputs to roughly 5 seconds. On this platform, you can select 5, 10, or 15-second durations for your generated videos.
What modes does Wan 2.7 support?
Wan 2.7 offers text-to-video, image-to-video, first/last frame video (FLF2V), and instruction-based editing tools. For this specific page, only text-to-video and image-to-video are active and available to use.
What resolutions does Wan 2.7 support?
Wan 2.7 supports output resolutions of 480P, 720P, and 1080P. You can generate content in both 16:9 landscape and 9:16 portrait aspect ratios for flexible use cases.
Is Wan 2.7 open source?
Wan 2.1 was fully open-sourced under the Apache-2.0 license. At launch, Wan 2.7’s official open-source release status had not been confirmed. For the most up-to-date details, visit the Alibaba Wan GitHub repository at github.com/Wan-Video.
How does Wan 2.7 compare to Wan 2.6?
Wan 2.7 introduces several key upgrades over Wan 2.6: first/last frame control, 9-grid multi-image input, support for up to 5 reference video inputs, and instruction-based editing, all of which were absent in the prior model. It also boosts maximum clip length to 15 seconds, plus improved motion physics and better character consistency across generated clips.
Still have unanswered questions? Our dedicated support team is ready to assist you