Wan 3.0 is an AI video generator for text-to-video, image-to-video, and multimodal creation. Generate from text, images, video clips, and audio, edit existing footage, or extend a scene. A normal generation runs 2–30 seconds; an extension keeps the input clip and the new footage within 30 seconds together. On KreadoAI, Wan 3.0 Standard and Wan 3.0 Prime stay on this page: both support the same creative workflow, and Prime is the speed-optimized version.
Use up to 20 assets in one request: up to 10 images, 5 videos totaling 15 seconds, and 5 audio clips totaling 15 seconds. Cite them as Image 1, Video 1, and Audio 1 to help guide who appears, where they are, how they move, and what is heard. Uploads on KreadoAI follow the model picker
Wan 3.0 video editing follows a text instruction to add, remove, or replace visuals, and to adjust style, lighting, and effects. A reference image can guide the change
Wan 3.0 video extension can continue a shot forward, backward, or both ways while trying to preserve the original scene and visual direction. The input clip and the generated extension together must stay within 30 seconds
Standard (wan3.0-video) is the regular generation experience. Prime (wan3.0-video-prime) is optimized for speed, not a guaranteed quality step up. Both support 480P, 720P, and 1080P, 2–30 seconds, 30 FPS, MP4, and audio. On KreadoAI, Standard is 6 / 12 / 24 K credits per second and Prime is 9 / 18 / 36 K credits per second for 480P / 720P / 1080P. Live rates follow the picker
Start a text-to-video brief, add image, video, and audio references, or upload a clip to edit or extend. Point the prompt at Image 1, Video 1, or Audio 1 when you need a specific asset to lead
Text-to-video, image-to-video, and other AI video models for campaigns that need more than one generation style
Combine image, video, and audio references, edit a clip, or extend a scene with Wan 3.0 Standard or Prime