Meituan open-sources LongCat Video Avatar 1.6: 10+ min consistent avatar videos under MIT license
AdinaYakup · x · 2026-11-17
Meituan's LongCat team released LongCat-Video-Avatar-1.6, an open-source audio-driven human video generation model now on Hugging Face under an MIT license.
Key upgrades:
- Supports 10+ minute long-form video generation, mitigating error accumulation in autoregressive continuation while preserving identity and temporal coherence
- 8-step fast inference with end-to-end pipeline optimizations for efficient deployment
- More natural lip animation and lip-audio sync, reduced exaggerated mouth movements and lip coloration artifacts, better coordinated facial/body motion
The team reports human evaluation wins over 1.5 and four commercial systems, with a technical report available.
More from Multimodal
- Google shows off Gemini 4 Hydrogen with visual ASMR demo, available to all users — GeminiApp · 2026-11-17
- Bria open-sources RMBG 3.0 background removal model, ships MCP server and Claude skill — victormustar · 2026-11-17
- Same Prompt, Two Groks: Grok Chat vs Grok Imagine Video Generation Compared — minchoi · 2026-11-17
- Seedance 2.5 Turns Grey 3D Blockouts Into Finished Anime Shots—No MoCap Needed — minchoi · 2026-11-17
- A World Series Home Run Rebuilt as a Gaussian Splat, Watchable From the Catcher's Eyes — minchoi · 2026-11-17
- Full Room Gaussian Splat Relit in Blender 5.3 With Cycles Using Custom Delighting — minchoi · 2026-11-17