mk1-project / quickreduce
QuickReduce is a performant all-reduce library designed for AMD ROCm that supports inline compression.
☆24Updated last month
Alternatives and similar repositories for quickreduce:
Users that are interested in quickreduce are comparing it to the libraries listed below
- ☆68Updated 4 months ago
- Fast low-bit matmul kernels in Triton☆294Updated this week
- extensible collectives library in triton☆85Updated 3 weeks ago
- Applied AI experiments and examples for PyTorch☆262Updated last month
- ☆198Updated 9 months ago
- ☆98Updated last month
- NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the …☆150Updated last week
- Development repository for the Triton language and compiler☆118Updated this week
- ☆78Updated 5 months ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆106Updated 9 months ago
- Ahead of Time (AOT) Triton Math Library☆58Updated this week
- OpenAI Triton backend for Intel® GPUs☆182Updated this week
- ☆102Updated last month
- KernelBench: Can LLMs Write GPU Kernels? - Benchmark with Torch -> CUDA problems☆274Updated last week
- Collection of kernels written in Triton language☆120Updated 3 weeks ago
- Fastest kernels written from scratch☆245Updated 3 weeks ago
- Shared Middle-Layer for Triton Compilation☆246Updated last week
- Cataloging released Triton kernels.☆217Updated 3 months ago
- PyTorch emulation library for Microscaling (MX)-compatible data formats☆220Updated last week
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆248Updated 6 months ago
- ☆95Updated last year
- This repository contains the experimental PyTorch native float8 training UX☆224Updated 8 months ago
- A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")☆321Updated this week
- ☆201Updated this week
- Perplexity GPU Kernels☆251Updated this week
- MLIR-based partitioning system☆80Updated this week
- Fast Hadamard transform in CUDA, with a PyTorch interface☆174Updated 11 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆82Updated last week
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆116Updated this week
- A Vectorized N:M Format for Unleashing the Power of Sparse Tensor Cores☆51Updated last year