ByteDance Lumina releases Seedance 2.5: 30s video gen & 50 reference controls
umesh_ai · x · 2026-08-17
ByteDance's Lumina launches Seedance 2.5, an AI video generator designed for industrial-scale creation. Key features include:
- 30-Second Continuous Output: Generates single clips up to 30 seconds with consistent camera movement and emotion, removing the need for post-production stitching.
- 50 Full-Modal References: Supports up to 50 inputs (images, video, audio) to precisely control characters, scenes, and camera movements.
- Precise Local Editing: Allows modification of specific elements (background, expressions) while preserving the overall composition and motion.
- Multilingual Support: Offers creative controls in over 10 languages.
The update shifts focus from generating cool shots to directing narrative scenes.
More from Multimodal
- FrankenSim builds a physics-based virtual orchestra, piano and drum demos live — doodlestein · 2026-10-16
- vLLM officially supports Muse Spark Open: multimodal, 1M context, agentic — vllm_project · 2026-10-16
- InfoVis: 9B unified model tops open-source slide generation and editing, trained on 13M records — Jiahao Wang · 2026-10-16
- RepForcing: feature-space supervision improves few-step autoregressive video distillation — zju · 2026-10-16
- SCOPE: semantic span-aware GRPO objective improves T2I prompt rewriting — university-of-rochester · 2026-10-16
- Orbia benchmark asks: do generated videos form a consistent 3D world? — tencent · 2026-10-16