Skip to content
Change the repository type filter

All

    Repositories list

    • ms-swift

      Public
      Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3…
      Python
      Apache License 2.0
      1.6k003Updated Jul 31, 2026Jul 31, 2026
    • Ongoing research training transformer models at scale
      Python
      Other
      4.3k002Updated Jul 31, 2026Jul 31, 2026
    • MoonEP

      Public
      MoonEP: A Perfectly Balanced Expert Parallelism Library via Dynamic Redundant Experts
      Python
      MIT License
      102000Updated Jul 31, 2026Jul 31, 2026
    • 🚀 Efficient implementations of state-of-the-art linear attention models
      Python
      MIT License
      624100Updated Jul 31, 2026Jul 31, 2026
    • TeraMoE

      Public
      TeraMoE: A cross-node expert-parallel MoE training library that uses a cooperative persistent kernel to overlap dispatch, expert compute, and combine.
      Cuda
      MIT License
      0200Updated Jul 31, 2026Jul 31, 2026
    • Python
      491002Updated Jul 29, 2026Jul 29, 2026
    • FlashKDA

      Public
      FlashKDA: high-performance Kimi Delta Attention kernels
      Cuda
      MIT License
      104000Updated Jul 29, 2026Jul 29, 2026
    • cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.
      Python
      MIT License
      231000Updated Jul 28, 2026Jul 28, 2026
    • An extended sonic-moe implementation, with FP8 support fully developed by agents
      Python
      Apache License 2.0
      97208Updated Jul 27, 2026Jul 27, 2026
    • PaddlePaddle Contributor Stories
      HTML
      1000Updated Jul 27, 2026Jul 27, 2026
    • blog

      Public
      PFCC 社区博客
      Vue
      Creative Commons Zero v1.0 Universal
      291413Updated Jul 27, 2026Jul 27, 2026
    • dochooks

      Public
      Python
      MIT License
      0210Updated Jul 27, 2026Jul 27, 2026
    • ast-grep pre-commit hook that uses the PyPI distribution for seamless integration with Python projects.
      Python
      MIT License
      0000Updated Jul 24, 2026Jul 24, 2026
    • DeepGEMM

      Public
      DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
      Cuda
      MIT License
      1.1k003Updated Jul 24, 2026Jul 24, 2026
    • Training library for Megatron-based models with bidirectional Hugging Face conversion capability
      Python
      Apache License 2.0
      436000Updated Jul 14, 2026Jul 14, 2026
    • ci-bypass

      Public
      Bypass CI checks for GitHub Actions
      TypeScript
      MIT License
      2110Updated Jul 12, 2026Jul 12, 2026
    • Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
      Python
      Other
      665400Updated Jul 11, 2026Jul 11, 2026
    • Python
      8000Updated Jul 1, 2026Jul 1, 2026
    • C++
      40011Updated Jun 29, 2026Jun 29, 2026
    • 基于PaddlePaddle的自动化功能开发小组
      Python
      Apache License 2.0
      471121Updated Jun 24, 2026Jun 24, 2026
    • Fast Hadamard transform in CUDA, with a PyTorch interface
      C
      BSD 3-Clause "New" or "Revised" License
      68000Updated Jun 8, 2026Jun 8, 2026
    • gloo

      Public
      Collective communications library with various primitives for multi-machine training.
      C++
      Other
      363000Updated Jun 4, 2026Jun 4, 2026
    • FlashMLA

      Public
      FlashMLA: Efficient Multi-head Latent Attention Kernels
      C++
      MIT License
      1.1k001Updated Jun 2, 2026Jun 2, 2026
    • Camp

      Public
      飞桨护航计划集训营
      Mermaid
      10820350Updated May 25, 2026May 25, 2026
    • FlashInfer: Kernel Library for LLM Serving
      Cuda
      Apache License 2.0
      1.2k004Updated May 19, 2026May 19, 2026
    • DeepEP

      Public
      DeepEP: an efficient expert-parallel communication library
      Cuda
      MIT License
      1.4k003Updated May 19, 2026May 19, 2026
    • quack

      Public
      A Quirky Assortment of CuTe Kernels
      Python
      Apache License 2.0
      148001Updated May 12, 2026May 12, 2026
    • PPOCRLabelv3 is a semi-automatic graphic annotation tool suitable for OCR field, with built-in PP-OCR model to automatically detect and re-recognize data.
      Python
      104433132Updated Apr 24, 2026Apr 24, 2026
    • Mooncake

      Public
      Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
      C++
      Apache License 2.0
      1k000Updated Apr 23, 2026Apr 23, 2026
    • PyTorch media decoding and encoding
      Python
      BSD 3-Clause "New" or "Revised" License
      113001Updated Apr 22, 2026Apr 22, 2026
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.