NVIDIA / compute-evalLinks
Evaluating Large Language Models for CUDA Code Generation ComputeEval is a framework designed to generate and evaluate CUDA code from Large Language Models.
☆74Updated last month
Alternatives and similar repositories for compute-eval
Users that are interested in compute-eval are comparing it to the libraries listed below
Sorting:
- TritonParse: A Compiler Tracer, Visualizer, and Reproducer for Triton Kernels☆171Updated last week
- How to ensure correctness and ship LLM generated kernels in PyTorch☆121Updated last week
- AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming☆111Updated last week
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆385Updated last week
- Official Problem Sets / Reference Kernels for the GPU MODE Leaderboard!☆158Updated last week
- ☆250Updated this week
- High-Performance SGEMM on CUDA devices☆112Updated 10 months ago
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆286Updated this week
- Efficient implementation of DeepSeek Ops (Blockwise FP8 GEMM, MoE, and MLA) for AMD Instinct MI300X☆71Updated 2 weeks ago
- Write a fast kernel and run it on Discord. See how you compare against the best!☆61Updated last week
- Applied AI experiments and examples for PyTorch☆305Updated 3 months ago
- Fast low-bit matmul kernels in Triton☆398Updated this week
- ☆93Updated last year
- Collection of kernels written in Triton language☆167Updated 7 months ago
- An experimental CPU backend for Triton☆164Updated last week
- Cataloging released Triton kernels.☆267Updated 2 months ago
- A curated collection of resources, tutorials, and best practices for learning and mastering NVIDIA CUTLASS☆244Updated 6 months ago
- extensible collectives library in triton☆91Updated 7 months ago
- This repository contains companion software for the Colfax Research paper "Categorical Foundations for CuTe Layouts".☆77Updated last month
- torchcomms: a modern PyTorch communications API☆291Updated this week
- Perplexity GPU Kernels☆529Updated 2 weeks ago
- A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.☆633Updated this week
- Fastest kernels written from scratch☆394Updated 2 months ago
- kernels, of the mega variety☆608Updated last month
- KernelBench: Can LLMs Write GPU Kernels? - Benchmark with Torch -> CUDA (+ more DSLs)☆663Updated last week
- ring-attention experiments☆155Updated last year
- A Quirky Assortment of CuTe Kernels☆660Updated 3 weeks ago
- Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.☆401Updated this week
- ☆148Updated 10 months ago
- AI Tensor Engine for ROCm☆301Updated this week