30-Second Video Generation
Create up to 30 seconds of video in a single generation. Build richer scenes, continuous camera movements, and more complete visual stories without stitching together endless short clips.
Create cinematic videos from text, images, audio, video, and documents — with up to 30-second generation, flexible editing, and lower production costs.
Create longer stories, use more types of input, and keep your creative direction consistent from the first frame to the final cut.
Create up to 30 seconds of video in a single generation. Build richer scenes, continuous camera movements, and more complete visual stories without stitching together endless short clips.
Start with more than a text prompt. Bring in images, audio, videos, webpages, presentations, spreadsheets, PDFs, and other documents to turn existing content into engaging video.
Keep your creative direction connected across every scene. Use reference inputs to guide characters, products, spaces, visual styles, and key storytelling details with greater consistency.
Create natural-looking people with expressive faces, subtle micro-expressions, and believable body language. Bring products, environments, interfaces, and motion graphics to life with greater visual fidelity.
Change visuals, story details, and dialogue without rebuilding the entire video from scratch. Refine your ideas, test new directions, and turn every generation into a flexible creative draft.
Turn a reference video and a character image into a precise, expressive motion-controlled video in three simple steps.

Add a character image and a reference video with the movement you want to recreate.
Use a prompt to guide the background, lighting, camera direction, and visual style.
Create your video, review the result, and adjust your references or prompt for the next take.
Everything you need to know about creating controlled character motion with Wan 3.0.
Wan 3.0 is a next-generation multimodal AI video model that creates videos from text, images, audio, video, webpages, and documents. It is designed for longer storytelling, stronger reference consistency, and flexible video editing.
Wan 3.0 helps you create more content with fewer production steps. Generate longer videos in a single pass, reuse your existing assets, and refine your results without starting over. You can also explore ideas at a lower resolution first, then create the final version at a higher resolution when the concept is ready.
You can use text, images, audio, videos, webpages, and supported documents such as presentations, PDFs, spreadsheets, text files, and Markdown files. Turn your existing content into the starting point for a new video.
Wan 3.0 can generate videos up to 30 seconds in a single generation. It also supports smart duration recommendations and video extension, making it easier to develop longer and more complete creative ideas.
Wan 3 is designed to preserve important details across generated videos, including character appearance, clothing, accessories, product structure, spatial relationships, and visual style when suitable reference inputs are provided. You can also revise visual elements, story details, and dialogue without rebuilding the entire video from scratch.
Buzzy brings Wan 3.0 into a simple, creator-friendly workflow. Upload your assets, write a prompt, choose your resolution, aspect ratio, and duration, then generate your video in one streamlined workspace — without managing a separate API setup.
Buzzy provides free trial credits so you can explore Wan 3.0 and experience its core capabilities. If you need more generations, you can subscribe to unlock 7-day unlimited access to Wan 3.0, along with other premium models such as Seedance 2.5 and MiniMax H3.
Turn any idea, asset, or document into a cinematic video in minutes. Create more content with fewer production steps.