Club&Lab for PaddlePaddle contributors
Repositories list
72 repositories
blog
PublicPFCC 社区博客dochooks
PublicPaddleAPITest
Publictypos-pre-commit-mirror
PublicMegatron-LM
PublicDeepSelect
PublicTeraMoE
Publicsupersonic-moe
PublicAn extended sonic-moe implementation, with FP8 support fully developed by agentsDeepGEMM
PublicDeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scalingDeepEP
PublicDeepEP: an efficient expert-parallel communication libraryms-swift
PublicFlashMLA
PublicFlashMLA: Efficient Multi-head Latent Attention Kernelspfcclab.github.io
Publiccudnn-frontend
PubliccuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.flash-linear-attention
Public🚀 Efficient implementations of state-of-the-art linear attention modelsmcore-bridge
PublicPPOCRLabel
Publicci-bypass
Publicquack
PublicA Quirky Assortment of CuTe KernelsMoonEP
PublicMoonEP: A Perfectly Balanced Expert Parallelism Library via Dynamic Redundant Expertstilelang-paddle
PublicDomain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernelsFlashKDA
PublicFlashKDA: high-performance Kimi Delta Attention kernelsMegatron-Bridge
PublicPaddleCppAPITest
PublicPaddleAutoProject
Publicfast-hadamard-transform
PublicFast Hadamard transform in CUDA, with a PyTorch interfacegloo
PublicCamp
Publicflashinfer
PublicFlashInfer: Kernel Library for LLM Serving
ProTip! When viewing an organization's repositories, you can use the
props. filter to filter by custom property.