minefantast

minefantast's Stars

• zjhellofss/KuiperLLama

  校招、秋招、春招、实习好项目，带你从零动手实现支持LLama2/3和Qwen2.5的大模型推理框架。

  Language:C++22850

• DefTruth/CUDA-Learn-Notes

  📚Modern CUDA Learn Notes with PyTorch: Tensor/CUDA Cores, 📖150+ CUDA Kernels with PyTorch bindings, 📖HGEMM/SGEMM (95%~99% cuBLAS performance), 📖100+ LLM/CUDA Blogs.

  Language:Cuda1.5k161

• daemyung/metal-by-tutorials-2nd

  Metal by Tutorials By the raywenderlich Tutorial Team

  Language:Swift367

• Yinghan-Li/YHs_Sample

  Yinghan's Code Sample

  Language:Cuda28954

• KhronosGroup/Vulkan-Samples

  One stop solution for all Vulkan samples

  Language:C++4.3k649

• Bruce-Lee-LY/decoding_attention

  Decoding Attention is specially optimized for multi head attention (MHA) using CUDA core for the decoding stage of LLM inference.

  Language:C++241

• Bruce-Lee-LY/flash_attention_inference

  Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.

  Language:C++293

• Bruce-Lee-LY/cuda_hgemm

  Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruction.

  Language:Cuda30267

• leimao/TensorRT-Custom-Plugin-Example

  Quick and Self-Contained TensorRT Custom Plugin Implementation and Integration

  Language:C++388

• leimao/CUDA-GEMM-Optimization

  CUDA Matrix Multiplication Optimization

  Language:Cuda14112

• openai/openai-gemm

  Open single and half precision gemm implementations

  Language:C37485

• KhronosGroup/glslang

  Khronos-reference front end for GLSL/ESSL, partial front end for HLSL, and a SPIR-V generator.

  Language:C++3.1k842

• Keenuts/vulkan-compute

  related to virglrender-vulkan: basic compute test application

  Language:C122

• vblanco20-1/vulkan-guide

  Introductory guide to vulkan.

  Language:SCSS948220

• KomputeProject/kompute

  General purpose GPU compute framework built on Vulkan to support 1000s of cross vendor graphics cards (AMD, Qualcomm, NVIDIA & friends). Blazing fast, mobile-enabled, asynchronous and optimized for advanced GPU data processing usecases. Backed by the Linux Foundation.

  Language:C++2k155

• NVIDIA/TransformerEngine

  A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper and Ada GPUs, to provide better performance with lower memory utilization in both training and inference.

  Language:Python2k328

• Sunt-ing/stick

  :innocent: A PyTorch-like deep learning framework. Just for fun.

  Language:Python1368

• feifeibear/LLMSpeculativeSampling

  Fast inference from large lauguage models via speculative decoding

  Language:Python57157

• tpoisonooo/how-to-optimize-gemm

  row-major matmul optimization

  Language:C++59180

• NervanaSystems/maxas

  Assembler for NVIDIA Maxwell architecture

  Language:Sass950162

• cloudcores/CuAssembler

  An unofficial cuda assembler, for all generations of SASS, hopefully ：）

  Language:Python40572

• minitorch/minitorch

  The full minitorch student suite.

  Language:Python1.9k406

• GetUpEarlier/minit

  Language:Python231

• karpathy/llm.c

  LLM training in simple, raw C/CUDA

  Language:Cuda24.5k2.8k

• Jittor/jittor

  Jittor is a high-performance deep learning framework based on JIT compiling and meta-operators.

  Language:Python3.1k313

• mlc-ai/notebooks

  Language:Jupyter Notebook18964

• mlc-ai/mlc-zh

  Language:Python59064

• hyperai/tvm-cn

  TVM Documentation in Chinese Simplified / TVM 中文文档

  Language:TypeScript957155

• alibaba/MNN

  MNN is a blazing fast, lightweight deep learning framework, battle-tested by business-critical use cases in Alibaba

  Language:C++8.7k1.7k

• ggerganov/llama.cpp

  LLM inference in C/C++

  Language:C++68.1k9.8k