ByteDance's Seedance 2.5 Generates 30-Second Videos in One Go: Hands-On
卡尔的AI沃茨 · wechat · 2026-07-31
ByteDance has released Seedance 2.5, its latest AI video generation model. The new version increases single-shot generation time to 30 seconds and supports up to 50 reference materials, enabling coherent short films with complete narrative arcs. Tests show significant improvements in character consistency, realistic facial details (removing the previous waxy look), physics, and multilingual lip-syncing. Additionally, the paired platform introduces a 3D director stage for pre-visualizing camera movements and a partial re-shoot feature, greatly reducing modification costs and lowering production barriers.
More from Multimodal
- Nous Research Opens Free FLUX 3 Preview Access to All Users — robrombach · 2026-08-01
- Gaussian Splatting Generates Volumetric Fog Without Any Proxy Meshes — willeastcott · 2026-08-01
- MiniMax H3 ComfyUI Workflow Now Available — EverythingMacPro · 2026-08-01
- Generative Feedback Loop: AI 'Digests Itself' Over 100 Iterations — mars_santa · 2026-08-01
- Fully Local Generation: Creating a Metal Music Video with LTX 2.3 and Suno — Luzifee-666 · 2026-07-31
- a16z to host AI creative tools meetup in SF next week — venturetwins · 2026-07-31