OpenMotionLab/MotionGPT3
MotionGPT3: Human Motion as a Second Modality, a MoT-based framework for unified motion understanding and generation
[NeurIPS 2023] MotionGPT: Human Motion as a Foreign Language, a unified motion-language generation model using LLMs
$ git clone https://github.com/OpenMotionLab/MotionGPT.gitMotionGPT3: Human Motion as a Second Modality, a MoT-based framework for unified motion understanding and generation
[NeurIPS 2024] MeshXL: Neural Coordinate Field for Generative 3D Foundation Models, a 3D fundamental model for mesh generation
21 Lessons, Get Started Building with Generative AI
RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
Harness LLMs with Multi-Agent Programming
InternLM-XComposer2.5-OmniLive: A Comprehensive Multimodal System for Long-term Streaming Video and Audio Interactions
Data from GitHub ยท snapshot Sep 24, 2026