🚀 LongCat-Video Now Open-Source: Text/Image-to-Video + Video Continuation in One Model 🏆 Text/Image-to-Video Performance Hits Open-Source SOTA 🎬 Minutes-Long High-Quality Videos: No Color Drift/Quality Loss (Industry-Standout) ⚙ 13.6B Params | Strong Open-Source DiT-Based Unified Multitask Video Base Model ⚡ C2F Pipeline + Block Sparse Attention: 720p/30fps Video in Minutes 🤗 Open-Source Links: GitHub:github.com/meituan-longcat/L… Hugging Face:huggingface.co/meituan-longc… Project Page:meituan-longcat.github.io/Lo…

Oct 25, 2025 · 1:57 PM UTC

Replying to @Meituan_LongCat
Congrats, benchmarks look very promising! Thanks for sharing this with the world.
Love to see it! @Meituan_LongCat released an open video gen model that, according to benchmarks, plays in Wan 2.2 league. Interesting new capability: Video continuation, long coherent videos.
12
Replying to @Meituan_LongCat
Would be interesting to see a smaller model targeting consumer cards. That will help you unlock community adoption!
1
4
Replying to @Meituan_LongCat
brrr too many new models to try these last 2 weeks
1
4
Replying to @Meituan_LongCat
Thank you longcat 💚🐱
1
Replying to @Meituan_LongCat
美团也做大模型了啊😲
Replying to @Meituan_LongCat
Longcat!!
Replying to @Meituan_LongCat
Open-source text-to-video models are advancing rapidly.
Replying to @Meituan_LongCat
guys how can I become official tester for longcat models. GPU POOR!