licj15

llama3

The official Meta Llama 3 GitHub site

starter-workflows

Accelerating new GitHub Actions workflows

tiny-gpu

A minimal GPU design in Verilog to learn how GPUs work from the ground up

gpt-fast

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

cutlass

CUDA Templates for Linear Algebra Subroutines

arxiv-latex-cleaner

arXiv LaTeX Cleaner: Easily clean the LaTeX code of your paper to submit to arXiv

dust3r

DUSt3R: Geometric 3D Vision Made Easy

warp

A Python framework for high performance GPU simulation and graphics

FluidX3D

The fastest and most memory efficient lattice Boltzmann CFD software, running on all GPUs via OpenCL. Free for non-commercial use.

DeepSeek-V2

DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

pbrt-v4

Source code to pbrt, the ray tracer described in the forthcoming 4th edition of the "Physically Based Rendering: From Theory to Implementation" book.

Awesome-LLM-Inference

📖A curated list of Awesome LLM Inference Paper with codes, TensorRT-LLM, vLLM, streaming-llm, AWQ, SmoothQuant, WINT8/4, Continuous Batching, FlashAttention, PagedAttention etc.

LGM

[ECCV 2024 Oral] LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation.

cs249r_book

Collaborative book Machine Learning Systems

TensorRT-Model-Optimizer

TensorRT Model Optimizer is a unified library of state-of-the-art model optimization techniques such as quantization, sparsity, distillation, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM or TensorRT to optimize inference speed on NVIDIA GPUs.

tinymembench

Simple benchmark for memory throughput and latency

Grendel-GS

Ongoing research training gaussian splatting at scale by distributed system

owl

KIVI

KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache

nerfbaselines

Reproducible evaluation of NeRF methods

research-career-tools

SA-GS

ShiftAddLLM

ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization

Magicube

Magicube is a high-performance library for quantized sparse matrix operations (SpMM and SDDMM) of deep learning on Tensor Cores.

Edge-LLM

[DAC 2024] EDGE-LLM: Enabling Efficient Large Language Model Adaptation on Edge Devices via Layerwise Unified Compression and Adaptive Layer Tuning and Voting

ACT

[ICML 2024] Unveiling and Harnessing Hidden Attention Sinks: Enhancing Large Language Models without Training through Attention Calibration

mg-verilog

3D-Carbon

3D-Carbon: An Analytical Carbon Modeling Tool for 3D and 2.5D Integrated Circuits

ray-tracing-in-cuda

LogarithmicPosit

[DAC'24] Official Implementation of the Logarithmic Posit (LP) Number System