AwesomeRepos
Sign in

Tags / Graphics and 3D /

GPU compute

Library or tool for general-purpose GPU computation with CUDA, OpenCL, or compute shaders.

Repositories

14

Net stars gained, last 7 days

+384

Top-level Tag

Graphics and 3D

Net stars gained each day, last 90 days

2026-10-10: +172 net stars. Most from lonewolf0622/HelixSR, +150.2026-10-13: +189 net stars. Most from vosen/ZLUDA, +91.2026-10-20: +209 net stars. Most from T-Crypt/berth, +89.

Peak days

  1. 1

    Oct 10, 2026 +172 net stars (median day +36)

  2. 2

    Oct 13, 2026 +189 net stars (median day +36)

  3. 3

    Oct 20, 2026 +209 net stars (median day +36)

A peak day has at least 2 times the net stars of the median day of this chart, and at least 100 more. For each peak day, the list shows the repositories that gained the most net stars that day. It also shows their GitHub releases of that day or the day before (UTC), and their creation date when it was 14 days or less before. These events occur near the same day. They do not show why the stars came.

Largest

Most stars in total now.

  1. taichi-dev/taichi28K

    description: Productive, portable, and performant GPU programming in Python.

  2. triton-lang/triton21K

    readme: The aim of Triton is to provide an open-source environment to write fast code at higher productivity than CUDA, but also with higher flexibility than other existing DSLs.

  3. gpujs/gpu.js16K

    readme: GPU.js is a JavaScript Acceleration library for GPGPU (General purpose computing on GPUs) in JavaScript for Web and Node.

  4. vosen/ZLUDA15K

    readme: ZLUDA allows running unmodified CUDA applications using non-NVIDIA GPUs with near-native performance

  5. deepseek-ai/FlashMLA13K

    readme: FlashMLA is DeepSeek's library of optimized attention kernels, powering inference of the DeepSeek-V4.1 model on NVIDIA GPUs and Huawei Ascend NPUs.

  6. cupy/cupy12K

    readme: You can pass `ndarray` to existing CUDA C/C++ programs via RawKernels, use Streams for performance, or even call CUDA Runtime APIs directly.

  7. NVIDIA/cutlass11K

    readme: These are Python native interfaces for writing high-performance CUDA kernels based on core CUTLASS and CuTe concepts without any performance compromises.

  8. NVIDIA/thrustarchived · 5K

    readme: Thrust's **high-level** interface greatly enhances programmer **productivity** while enabling performance portability between GPUs and multicore CPUs.

  9. arrayfire/arrayfire4.9K

    description: ArrayFire: a general purpose GPU library.

  10. Sergio0694/ComputeSharp3.2K

    description: A .NET library to run C# code in parallel on the GPU through DX12, D2D1, and dynamically generated HLSL compute and pixel shaders

Newest Admissions

Entered the index by the rising rule, newest first.

  1. T-Crypt/berth286

    readme: Berth makes the programs on your PC take turns with the GPU's memory, so they stop crashing each other, and tells you who is holding it.

  2. lonewolf0622/HelixSR1K

    readme: HelixSR runs NVIDIA's DLSS Model E network as standard Direct3D 12 compute shaders, **without NVIDIA Tensor Cores, an NVIDIA driver, CUDA, or ROCm**.

All 14 in the Repository Index

How a repository gets this Tag

An agent reads the description, the GitHub topics, and the start of the README of each repository. It chooses Tags only from the Tag list in the AwesomeRepos source code. For each Tag, it quotes the text. A code check keeps the Tag only when the quote is in that text. The quote under each repository is that evidence. A Tag does not rank repositories. It does not say that one repository can replace another. To report a wrong Tag, use "Report a problem with this page" on the page of the repository. All numbers on this page count only visible repositories.