Stars
PyTorch code and models for VJEPA2 self-supervised learning from video.
Github Pages template based upon HTML and Markdown for personal, portfolio-based websites.
MAGI-1: Autoregressive Video Generation at Scale
Official inference framework for 1-bit LLMs
About Awesome things towards foundation agents. Papers / Repos / Blogs / ...
Awesome Reasoning LLM Tutorial/Survey/Guide
Code for NeurIPS'24 paper 'Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization'
Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models
A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations
Open-Sora: Democratizing Efficient Video Production for All
PyTorch implementation of FractalGen https://arxiv.org/abs/2502.17437
Genome modeling and design across all domains of life
DeepEP: an efficient expert-parallel communication library
这是一个从头训练大语言模型的项目，包括预训练、微调和直接偏好优 8970 化，模型拥有1B参数，支持中英文。
Large Concept Models: Language modeling in a sentence representation space
🚀 Efficient implementations of state-of-the-art linear attention models in Torch and Triton
A More Fair and Comprehensive Comparison between KAN and MLP
Pytorch Lightning入门中文教程，转载请注明来源。（当初是写着玩的，建议看完MNIST这个例子再上手）
深度学习系统笔记，包含深度学习数学基础知识、神经网络基础部件详解、深度学习炼丹策略、模型压缩算法详解。
tiktoken is a fast BPE tokeniser for use with OpenAI's models.
An open-source framework for collaborative AI agents, enabling diverse, distributed agents to team up and tackle complex tasks through internet-like connectivity.