Vidu AIモデルシリーズ
Vidu is Shengshu Technology's video generation model that turns text or image prompts into high-quality, realistic videos with strong reference-to-video consistency and coherent motion.
すべてのViduモデル

vidu/viduq4-r2v
Vidu's reference-to-video model for a newly composed shot that keeps characters, style, and voices consistent across camera changes. It uses 1 to 15 reference images and up to three audio clips, in a 3 to 16 second clip from 540p through 4K, including 3:4 and 4:3 frames.

vidu/viduq4-i2v
Vidu's image-to-video model for extending one still into a short clip with dialogue, sound effects, and camera movement. The image stays the opening frame, in a 3 to 16 second clip from 540p through 4K.

vidu/viduq3-ad
Vidu's advertising model for short brand films with paced cuts and consistent product and character detail. It builds the clip from a screenplay and 1 to 14 reference images, at 1080p in 16:9 or 9:16.

vidu/viduq3-drama
Vidu's short-drama model for scripted episodes with dialogue, character placement, and film-style camera work. It plans the shots from a script and 1 to 14 reference images, in an 8 to 12 second 1080p clip at 16:9 or 9:16.

vidu/viduq3-turbo-r2v
Vidu's fast reference-to-video model for character clips and social drafts that should keep a subject recognizable. It uses 1 to 7 reference images in a 3 to 16 second clip at 540p, 720p, or 1080p.

vidu/viduq3-r2v
Vidu's reference-to-video model for character-driven scenes that follow a prompt while keeping the person recognizable. It uses 1 to 7 reference images in a 3 to 16 second clip at 540p, 720p, or 1080p.

vidu/viduq3-mix-r2v
Vidu's balanced reference-to-video model for a specific subject performing a new action, with a consistent look and synchronized sound. It uses 1 to 7 reference images in a 1 to 16 second clip at 720p or 1080p.

vidu/one-click-trending-replicate
Vidu's trend-replication model for social clips that follow a reference video's pacing, cuts, and style. It rebuilds that look with 1 to 7 of your own images, at 540p, 720p, or 1080p.

vidu/lip-sync
Vidu's lip-sync model for dubbing, dialogue fixes, and making a face in an existing video speak a new track. It reshapes the mouth to the supplied audio and can lock identity to a reference face.

vidu/motion-sync
Vidu's motion-transfer model for dance clips, pet moves, and action copied onto another character. It applies the body movement from a reference video to the person or creature in your image.

vidu/one-click-ad-film
Vidu's one-click ad model for product showcases and promotional clips with narration. It turns 1 to 7 product images into a 10 to 60 second ad in Chinese or English.

vidu/one-click-general-film
Vidu's one-click film model for montages, travel pieces, and photo albums turned into a finished cut. It assembles 1 to 7 images into a 10 to 180 second film.

vidu/viduq2-pro-multi-frame
Vidu's multi-frame model for detailed storyboards and pose sequences where each supplied image should appear in order. It animates a start frame plus up to eight more frames, at 540p, 720p, or 1080p.

vidu/viduq2-turbo-multi-frame
Vidu's fast multi-frame model for storyboards and pose sequences that should pass through specific images. It animates a start frame plus up to eight more frames, at 540p, 720p, or 1080p.

vidu/viduq2-pro-digital-human
Vidu's portrait model for spokespersons, lessons, and character close-ups that need realistic expression. It animates one portrait and can match the mouth to a supplied audio track, at 540p, 720p, or 1080p.

vidu/viduq2-turbo-digital-human
Vidu's fast portrait model for talking-head clips, virtual presenters, and avatars. It animates one portrait and can match the mouth to a supplied audio track, at 540p, 720p, or 1080p.

vidu/template-story
Vidu's template story model for themed short films that place your characters into a ready-made plot, such as a love story, a workday, or Journey to the West. It builds the film from your character images and the chosen template.

vidu/viduq3-pro-fl2v
Vidu's cinematic first-and-last-frame model for commercial transitions and storyboard shots that need a clear path between two images. It bridges the start and end frames in a 1 to 16 second clip at 540p, 720p, or 1080p.

vidu/viduq3-turbo-fl2v
Vidu's fast first-and-last-frame model for morphs, before-and-after shots, and storyboard fills. It bridges a start image and an end image in a 1 to 16 second clip at 540p, 720p, or 1080p.

vidu/viduq3-turbo-i2v
Vidu's faster image-to-video model for social clips, animatics, and quick iterations from one still, with synchronized picture and sound. It makes a 1 to 16 second clip at 540p, 720p, or 1080p.

vidu/viduq3-pro-i2v
Vidu's cinematic image-to-video model for concept art, commercials, and stills that need stable characters, complex motion, and synchronized sound. It animates one opening frame into a 1 to 16 second clip at 540p, 720p, or 1080p.

vidu/viduq3-pro-fast-i2v
Vidu's fast image-to-video model for prototypes, product shots, and quick motion from a single still, with picture and sound together. It animates that frame into a 1 to 16 second clip at 720p or 1080p.

vidu/viduq3-pro-t2v
Vidu's cinematic text-to-video model for short films, ads, and concept scenes, with synchronized sound and automatic shot changes. It makes a 1 to 16 second clip at 540p, 720p, or 1080p, including vertical 9:16 and 3:4 frames.

vidu/viduq3-turbo-t2v
Vidu's fast text-to-video model for drafts, b-roll, and quick scene ideas, with synchronized picture and sound. It makes a 1 to 16 second clip at 540p, 720p, or 1080p, including vertical 9:16 and 3:4 frames.




