Vidu S2, the world-leading real-time interactive model
Millisecond response: the interaction and the picture happen at once
Vidu S2-Avatar
Natively full-duplex audio-visual avatar: sound, picture and timing understood together, so it listens, watches and speaks at once — as live conversation or offline batch generation
Vidu S2-Editing
Streaming video editing for real production work: rewrite style, wardrobe and identity mid-playback while timing stays consistent and identity holds
초기 이미지 및 음성 커스터마이즈
초기 이미지(사람, 애니메이션 또는 반려동물)와 음성을 선택해 캐릭터를 만들 수 있습니다
Vidu S2, the world-leading real-time interactive model
Millisecond response: the interaction and the picture happen at once
Vidu S2-Avatar
Natively full-duplex audio-visual avatar: sound, picture and timing understood together, so it listens, watches and speaks at once — as live conversation or offline batch generation
Vidu S2-Editing
Streaming video editing for real production work: rewrite style, wardrobe and identity mid-playback while timing stays consistent and identity holds
초기 이미지 및 음성 커스터마이즈
초기 이미지(사람, 애니메이션 또는 반려동물)와 음성을 선택해 캐릭터를 만들 수 있습니다
Vidu S2, the world-leading real-time interactive model
Millisecond response: the interaction and the picture happen at once
Vidu S2-Avatar
Natively full-duplex audio-visual avatar: sound, picture and timing understood together, so it listens, watches and speaks at once — as live conversation or offline batch generation
Vidu S2-Editing
Streaming video editing for real production work: rewrite style, wardrobe and identity mid-playback while timing stays consistent and identity holds
초기 이미지 및 음성 커스터마이즈
초기 이미지(사람, 애니메이션 또는 반려동물)와 음성을 선택해 캐릭터를 만들 수 있습니다
What Vidu S Can Do
Vidu S2-Avatar
Natively full-duplex audio-visual avatar: sound, picture and timing understood together, so it listens, watches and speaks at once — as live conversation or offline batch generation
Vidu S2-Editing
Streaming video editing for real production work: rewrite style, wardrobe and identity mid-playback while timing stays consistent and identity holds
초기 이미지 및 음성 커스터마이즈
초기 이미지(사람, 애니메이션 또는 반려동물)와 음성을 선택해 캐릭터를 만들 수 있습니다
Model Family
Video Generation
Vidu R1
Vidu R1 is a next-generation AI video model built for ultra-high-quality, instruction-following generation, with marked gains in visual fidelity, inference speed and prompt adherence
Learn more
Real-time Interaction
Vidu S1
Vidu S1 is a world-leading real-time interactive model that streams interactive video, making conversation feel live rather than generated, built for tutoring, support and other two-way scenarios
Learn more
Latest
Vidu R1 is here: video generation at 4K
Output up to 4K, with gains across visual fidelity and instruction following
Model releaseLearn moreVidu S1 real-time interactive model enters open beta
Streaming generation at 540P and 25 FPS, with unlimited session length
Model releaseLearn moreMulti-image reference generation ships
Blend several reference images, with markedly better subject consistency
ProductLearn moreThe Vidu API platform opens
Batch video generation, with full documentation and samples
DevelopersLearn more
FAQs about Vidu R1
Vidu R1 is Vidu's next-generation AI video model, designed for ultra-high-quality, instruction-following video generation
R1 delivers significantly higher visual fidelity, faster inference, and more precise prompt adherence compared to Q3
Vidu R1 supports up to 4K resolution output
Yes, Vidu R1 is available on both desktop and mobile via the Vidu app
Select Vidu R1 as your model, enter a text prompt or upload a reference image, then generate and download