Cuda Python Packages

sglang

SGLang is a high-performance serving framework for large language models and multimodal models.

vllm

A high-throughput and memory-efficient inference and serving engine for LLMs

ctranslate2

Fast inference engine for Transformer models

catboost

A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU.

nvidia-cutlass-dsl

CUDA Templates and Python DSLs for High-Performance Linear Algebra

flashinfer-python

FlashInfer: Kernel Library for LLM Serving

nvidia-cutlass-dsl-libs-base

CUDA Templates and Python DSLs for High-Performance Linear Algebra

torchao

PyTorch native quantization and sparsity for training and inference

cupy-cuda12x

NumPy & SciPy for GPU

flashinfer-cubin

FlashInfer: Kernel Library for LLM Serving

cog

Containers for machine learning

open3d

Open3D: A Modern Library for 3D Data Processing

torchrec

Pytorch domain library for recommendation systems

warp-lang

A Python framework for GPU-accelerated simulation, robotics, and machine learning.

pennylane-lightning

The Lightning plugin ecosystem provides fast quantum state-vector and tensor network simulators written in C++ for use with PennyLane.

nvitop

An interactive NVIDIA-GPU process viewer and beyond, the one-stop solution for GPU process management.

cupy-cuda11x

NumPy & SciPy for GPU

cupy-cuda13x

NumPy & SciPy for GPU

sglang-kernel

SGLang is a high-performance serving framework for large language models and multimodal models.

libcudf-cu12

cuDF - GPU DataFrame Library

sgl-kernel

SGLang is a high-performance serving framework for large language models and multimodal models.

rmm-cu12

RAPIDS Memory Manager

pylibraft-cu12

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.

libraft-cu12

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.