LLM training in simple, raw C/CUDA

📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉

DeepEP: an efficient expert-parallel communication library

DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling

Tile primitives for speedy kernels

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

This package contains the original 2012 AlexNet code.

[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.

Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity

CPM.cu is a lightweight, high-performance CUDA implementation for LLMs, optimized for end-device inference and featuring cutting-edge techniques in sparse architecture, speculative sampling and qua…

A lightweight design for computation-communication overlap.

NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer

Benchmark code for the "Online normalizer calculation for softmax" paper

Batch computation of the linear assignment problem on GPU.

Source code for the CPU-Free model - a fully autonomous execution model for multi-GPU applications that completely excludes the involvement of the CPU beyond the initial kernel launch.