Top CUDA repositories

Biggest movers this week

git clone https://github.com/flashinfer-ai/flashinfer.git

git clone https://github.com/NVIDIA/warp.git

git clone https://github.com/wici-ai/inferssd.git

git clone https://github.com/debpalash/VoiceStudio.git

git clone https://github.com/XuehaiPan/nvitop.git

git clone https://github.com/sgl-project/sglang.git

Trending today

git clone https://github.com/debpalash/VoiceStudio.git

Top 30 on the Elo ladder

1. #10
   debpalash/VoiceStudio

   VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing…

   • Legendary #10

   2,550 Elo +40111kPython
2. #111
   vllm-project/vllm

   High-throughput LLM serving engine with PagedAttention and an OpenAI-compatible server.

   • Diamond I

   2,222 Elo −596.9kPython
3. #185
   jamiepine/voicebox

   The open-source AI voice studio. Clone, dictate, create.

   • Diamond I

   2,137 Elo −2060.7kPython
4. #275
   sgl-project/sglang

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

   • Diamond II

   2,080 Elo +3538.9kPython
5. #1,147
   hashcat/hashcat

   World's fastest and most advanced password recovery utility

   • Emerald III

   1,835 Elo −3327.3kC
6. #1,229
   bendlang/bend

   Bend 2: a fast language that blocks AI mistakes via proof. Install: curl -fsSL https://bend-lang.com/install.sh | sh

   • Emerald III

   1,823 Elo −3125kTypeScript
7. #1,860
   LMCache/LMCache

   LMCache: Supercharge Your LLM with the Fastest KV Cache Layer

   • Platinum I

   1,743 Elo +2712.3kPython
8. #1,865
   NVIDIA/TensorRT-LLM

   TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports…

   • Platinum I

   1,742 Elo −1915kPython
9. #2,593
   flashinfer-ai/flashinfer

   FlashInfer: Kernel Library for LLM Serving

   • Platinum IV

   1,668 Elo +816.9kCuda
10. #2,810
    tracel-ai/burn

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

    • Platinum IV

    1,651 Elo −1616.2kRust
11. #3,054
    isl-org/Open3D

    Open3D: A Modern Library for 3D Data Processing

    • Platinum V

    1,633 Elo +1114.2kC++
12. #3,426
    NVIDIA/cutlass

    CUDA Templates and Python DSLs for High-Performance Linear Algebra

    • Gold I

    1,609 Elo −1410.7kC++
13. #3,558
    vosen/ZLUDA

    CUDA on non-NVIDIA GPUs

    • Gold II

    1,601 Elo +2215.1kRust
14. #3,733
    NVlabs/instant-ngp

    Instant neural graphics primitives: lightning fast NeRF and more

    • Gold II

    1,589 Elo −117.6kCuda
15. #3,751
    NVIDIA/nvidia-docker

    Build and run Docker containers leveraging NVIDIA GPUs

    • Gold II

    1,588 Elo −3517.6k
16. #4,091
    NVIDIA/cuda-samples

    Samples for CUDA Developers which demonstrates features in CUDA Toolkit

    • Gold III

    1,568 Elo +239.7kC++
17. #4,292
    kaldi-asr/kaldi

    kaldi-asr/kaldi is the official location of the Kaldi project.

    • Gold IV

    1,557 Elo −4715.5kShell
18. #4,317
    xlite-dev/LeetCUDA

    An Open-source book with Modern CUDA learning Notes for Beginners - BF16/FP8/FP4, HGEMM, CuTe, Flash-Attention, etc.

    • Gold IV

    1,556 Elo +2012.1kTeX
19. #4,837
    NVIDIA/warp

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

    • Gold V

    1,527 Elo +727.3kPython
20. #5,249
    cupy/cupy

    NumPy & SciPy for GPU

    • Silver I

    1,508 Elo −2112.4kPython
21. #5,252
    numba/numba

    NumPy aware dynamic Python compiler using LLVM

    • Silver I

    1,507 Elo +1411.2kPython
22. #5,415
    srush/GPU-Puzzles

    Solve puzzles. Learn CUDA.

    • Silver II

    1,500 Elo −2512.5kJupyter Notebook
23. #5,881
    catboost/catboost

    A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification…

    • Silver III

    1,479 Elo +149.2kC++
24. #5,887
    NVIDIA/cudf

    cuDF - GPU DataFrame Library

    • Silver III

    1,478 Elo −239.8kC++
25. #7,738
    replicate/cog

    Containers for machine learning

    • Bronze II

    1,405 Elo −89.4kGo
26. #8,105
    Oneflow-Inc/oneflow

    OneFlow is a deep learning framework designed to be user-friendly, scalable and efficient.

    • Bronze III

    1,391 Elo −139.4kC++
27. #8,697
    hybridgroup/gocv

    Go package for computer vision using OpenCV 4 and beyond. Includes support for DNN, CUDA, OpenCV Contrib, and OpenVINO.

    • Bronze V

    1,366 Elo −117.5kGo
28. #9,203
    XuehaiPan/nvitop

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

    • Copper II

    1,338 Elo +367.2kPython
29. #9,822
    kroma-network/tachyon

    Modular ZK(Zero Knowledge) backend accelerated by GPU

    • Copper IV

    1,270 Elo −207.6kC++
30. #10,083
    wici-ai/inferssd

    Run models larger than your RAM or VRAM by streaming MoE experts from SSD. Built on llama.cpp.

    • Copper V

    838 Elo +721.1kC++