Specify whether the video consists of a single continuous shot or multiple switched shots.
Examples
Wan 2.6, 2.5 & Pro Cinematic AI Video Generator
The latest open-source cinematic video generation models from Alibaba Tongyi Wanxiang Team
Wan 2.6 is the next-generation cinematic video generation model from Alibaba Tongyi Wanxiang, pioneering multi-shot storytelling, native audio generation, and video style transfer. Combined with Wan 2.5's exceptional Image-to-Video capabilities, it offers a professional creation workflow from script to final cut.
Wan 2.6 Cinematic Showcase
Enjoy cinematic videos generated by Wan 2.6. Experience the smoothness of multi-shot storytelling, the immersion of audio-visual synchronization, and the artistic charm of style transfer, witnessing the infinite possibilities of AI video creation.
Latest Technology from Alibaba Tongyi Wanxiang
Wan 2.6 introduces breakthrough multi-shot storytelling and native audio generation technologies, elevating AI video generation from single shots to complete narratives. Together with the powerful foundation of Wan 2.5, it forms an industry-leading video generation matrix.
Full-Process Creation Mode
Supports Text-to-Video (T2V), Image-to-Video (I2V), multi-shot sequencing, and audio-visual synchronization. Wan 2.6 is optimized for long video coherence, generating up to 15 seconds of narrative video with automatically matched high-fidelity audio.
Cinematic Narrative Expression
Wan 2.6 focuses on cinematic language, featuring precise camera control (Pan, Tilt, Zoom) and lighting preservation. The new video reference feature allows users to replicate visual styles and pacing from reference videos, realizing director-level creative intent.
Wan 2.6 Core Features
Multi-Shot Storytelling
Wan 2.6 pioneers multi-shot generation, allowing a single video to contain multiple shots. The model automatically handles smooth transitions and logical coherence between shots, achieving true cinematic storytelling in a single generation.
Native Audio Generation
Introducing audio-visual integration, Wan 2.6 generates synchronized background music, ambient sounds, and Foley effects based on the video content, ensuring perfect audio-visual matching and eliminating the need for post-production dubbing.
Video Style Transfer (Video-to-Video)
By inputting a reference video, Wan 2.6 can learn and transfer its visual style, color tone, and camera pacing to new video content, providing creators with advanced artistic control and stylized creation capabilities.
Enhanced Text-to-Video (T2V)
Leveraging powerful language understanding, Wan 2.6 accurately parses complex scripts, rendering detail-rich scenes and character performances. It supports bilingual prompts (English/Chinese) and generates cinematic visuals up to 1080P resolution.
Create Professional Videos with Wan 2.6
Ideation & Input
Enter your creative script or upload reference images/videos. For Wan 2.6, try describing complex plots with multiple scene changes, or specify a particular background music style.
Configure Settings
Select generation mode (T2V/I2V). Set camera movements, aspect ratio, and potential duration. If using Wan 2.6, enable 'Auto Audio' for an integrated audio-visual experience.
Generate & Preview
Click generate, and the Alibaba Wan model will render your request. Once complete, preview the full video with visuals and sound, experiencing cinematic narrative effects.
Frequently Asked Questions about Wan 2.6
Learn more about the Alibaba Tongyi Wanxiang Wan 2.6 cinematic video generation model
What are the major upgrades in Wan 2.6 compared to previous versions?
The biggest upgrades in Wan 2.6 are the introduction of multi-shot storytelling and native audio generation, along with support for video style transfer. This elevates video generation from 'clips' to 'stories', with significant improvements in duration and quality.
What is the Multi-Shot Storytelling feature?
Multi-shot storytelling allows the model to present multiple different scenes or perspectives (shots) within a continuous video generation process and automatically handle the cuts between them, creating a complete segment with a cinematic feel rather than a single fixed shot.
Does Wan 2.6 generate audio?
Yes, Wan 2.6 supports native audio generation. It not only generates visuals but also automatically creates matching background music and sound effects (like footsteps, wind, etc.) based on the visual content, achieving true audio-visual synchronization.
Can I make long videos with Wan 2.6?
Wan 2.6 supports generating videos up to 15 seconds (or longer depending on configuration). Combined with its multi-shot capabilities, it is perfect for creating short films, ad teasers, or social media stories, offering much greater expressive power than previous few-second clips.
Will Wan 2.5 and 2.2 still be available?
Wan 2.5 and 2.2 remain as foundational models in the Wan series, continuing to excel in specific Image-to-Video or lightweight generation tasks. Wan 2.6 is the high-end flagship version of the series, offering more comprehensive professional features.
How do I use the Video Style Transfer feature?
Simply upload a reference video with a specific style (e.g., oil painting, cyberpunk, or specific film grading), and Wan 2.6 will learn its visual characteristics to generate your text description into a new video with that same style.