Shengshu releases ViduS2: real-time costume, scene and prop switching driven by chat
生数科技 · wechat · 2026-09-22
Shengshu (Vidu) has released ViduS2, a new real-time interactive and editing model that breaks the previous limitation of fixed visuals in its Empresses in the Palace interactive demo. A single sentence can now drive real-time visual changes: scene switching with natural lighting adaptation, real-time prop generation (e.g. a sword in hand), costume changes with synced accessories, and instant dialogue responses with coherent micro-expressions.
ViduS2 decomposes character fidelity into four verifiable dimensions: persona consistency across long multi-turn dialogues, stability over sessions extended from 1 minute to 2 hours without quality decay, layered emotional expression with configurable action libraries, and real-time appearance/scene changes mid-session without restart. The solution targets gaming, brand livestreams, cultural tourism and enterprise training, with lightweight SaaS/MaaS access and support for 100+ languages and dialects.
More from Multimodal
- Refining MiniMax H3 720p video to 2K on an 8GB GPU: a ComfyUI workflow discussion — ashendonep · 2026-09-23
- Tencent Hunyuan's Hy Image 3.5 preview lands in ComfyUI with +30% human eval win rate — TencentHunyuan · 2026-09-23
- Anyone can now generate videos like these — full prompt shared — umesh_ai · 2026-09-23
- FlovaAI launches year-long price lock for Seedance 2.5 at $0.044/sec in 720p — thetripathi58 · 2026-09-23
- Qwen-Image-2.1 4-step turbo LoRA generation demo — CodeMichaelD · 2026-09-23
- "No amount of graphics made a bad game good": devs on 3D generation in game design — rms80 · 2026-09-23