Fully open reproduction of DeepSeek-R1

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

This repository contains demos I made with the Transformers library by HuggingFace.

SGLang is a fast serving framework for large language models and vision language models.

Implementation of Microscaling data formats in SystemVerilog.

CUDA Learning guide

tpu-systolic-array-weight-stationary

Code for Neurips24 paper: QuaRot, an end-to-end 4-bit inference of large language models.

Material for gpu-mode lectures

FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.

OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.

Berkeley's Spatial Array Generator

Kernel Tuner

All in one vscode plugin for HDL development

Verilog AXI components for FPGA implementation

CUDA Templates for Linear Algebra Subroutines

A survey on Hardware Accelerated LLMs

Machine-Learning Accelerator System Exploration Tools

PyTorch emulation library for Microscaling (MX)-compatible data formats

the resources I use to learn computer science in my spare time

A modern Neovim configuration with full battery for Python, Lua, C++, Markdown, LaTeX, and more...

高颜值的第三方网易云播放器，支持 Windows / macOS / Linux

在vscode上的数字设计开发插件

Long Range Arena for Benchmarking Efficient Transformers

MLNLP: This repository is a collection of AI top conferences papers (e.g. ACL, EMNLP, NAACL, COLING, AAAI, IJCAI, ICLR, NeurIPS, and ICML) with open resource code

This repository contains source code to binarize any real-value word embeddings into binary vectors.