Stars
Implementing DeepSeek R1's GRPO algorithm from scratch
MoBA: Mixture of Block Attention for Long-Context LLMs
Distributed Triton for Parallel Systems
Official inference repo for FLUX.1 models
A high-performance distributed file system designed to address the challenges of AI training and inference workloads.
Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
Model Context Protocol Servers
DeepEP: an efficient expert-parallel communication library
FlashMLA: Efficient MLA decoding kernels
Distilabel is a framework for synthetic data and AI feedback for engineers who need fast, reliable and scalable pipelines based on verified research papers.
Fully open reproduction of DeepSeek-R1
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
The official repo of MiniMax-Text-01 and MiniMax-VL-01, large-language-model & vision-language-model based on Linear Attention
DSPy: The framework for programming—not prompting—language models
A beautiful, simple, clean, and responsive Jekyll theme for academics
The simplest, fastest repository for training/finetuning medium-sized GPTs.
A collection of LLM papers, blogs, and projects, with a focus on OpenAI o1 🍓 and reasoning techniques.
Tile primitives for speedy kernels
The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.
CUDA Python: Performance meets Productivity
Automatically block unwanted, leeches and abnormal BT peers with support for customized and cloud rules.| BT 反吸血工具 - 自动封禁不受欢迎、吸血和异常的 BT 客户端，并支持自定义规则。支持 qB/qBEE/Deluge/BiglyBT/BitComet
FlashInfer: Kernel Library for LLM Serving
A bibliography and survey of the papers surrounding o1
Graph-indexed Pandas DataFrames for analyzing hierarchical performance data