#cuda

LMCache/LMCache

Python

89

health

LMCache: Supercharge Your LLM with the Fastest KV Cache Layer

NVIDIA/TransformerEngine

Python

88

health

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.

tracel-ai/burn

Rust

88

health

Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

lupinemachines/lupine

C++

88

health

LUPINE is a GPU over IP bridge allowing GPUs on remote machines to be attached to CPU-only machines.

luigifcruz/CyberEther

C++

88

health

High-performance GPU-accelerated signal processing and visualization framework that runs anywhere.

flashinfer-ai/flashinfer

Python

87

health

FlashInfer: Kernel Library for LLM Serving

Avarok-Cybersecurity/atlas

Rust

87

health

Pure Rust Inference Engine

rapidsai/cudf

C++

86

health

cuDF - GPU DataFrame Library

rapidsai/cuml

Python

85

health

cuML - RAPIDS Machine Learning Library

kibae/onnxruntime-server

C++

85

health

ONNX Runtime Server: The ONNX Runtime Server is a server that provides TCP and HTTP/HTTPS REST APIs for ONNX inference.

parca-dev/parca-agent

Go

83

health

eBPF based always-on CPU/GPU profiler auto-discovering targets in Kubernetes and systemd, zero code changes or restarts needed!

Wangmerlyn/KeepGPU

Python

81

health

KeepGPU is a simple CLI app that keeps your GPUs running.

tenstorrent/tt-metal

C++

79

health

:metal: TT-NN operator library, and TT-Metalium low level kernel programming model.

quokka-astro/quokka

C++

76

health

Two-moment AMR radiation hydrodynamics (with self-gravity, particles, chemistry, and magnetic fields) on CPUs/GPUs for astrophysics

Seeed-Projects/reComputer-Jetson-for-Beginners

Python

69

health

Beginner's Guide to reComputer Jetson