tspeterkim / mixed-precision-from-scratchLinks
Mixed precision training from scratch with Tensors and CUDA
☆24Updated last year
Alternatives and similar repositories for mixed-precision-from-scratch
Users that are interested in mixed-precision-from-scratch are comparing it to the libraries listed below
Sorting:
- The simplest implementation of recent Sparse Attention patterns for efficient LLM inference.☆77Updated 2 weeks ago
- Experiment of using Tangent to autodiff triton☆79Updated last year
- ring-attention experiments☆144Updated 8 months ago
- ☆68Updated this week
- ☆78Updated 11 months ago
- Learn CUDA with PyTorch☆27Updated this week
- Code for studying the super weight in LLM☆107Updated 6 months ago
- ☆221Updated this week
- Cataloging released Triton kernels.☆238Updated 5 months ago
- A minimal implementation of vllm.☆44Updated 11 months ago
- Load compute kernels from the Hub☆191Updated last week
- CUDA and Triton implementations of Flash Attention with SoftmaxN.☆70Updated last year
- Boosting 4-bit inference kernels with 2:4 Sparsity☆79Updated 9 months ago
- The evaluation framework for training-free sparse attention in LLMs☆69Updated last week
- Flash-Muon: An Efficient Implementation of Muon Optimizer☆131Updated last week
- Write a fast kernel and run it on Discord. See how you compare against the best!☆46Updated this week
- Make triton easier☆46Updated last year
- Applied AI experiments and examples for PyTorch☆277Updated 3 weeks ago
- ☆13Updated 3 months ago
- Collection of kernels written in Triton language☆132Updated 2 months ago
- A bunch of kernels that might make stuff slower 😉☆51Updated this week
- PTX-Tutorial Written Purely By AIs (Deep Research of Openai and Claude 3.7)☆66Updated 3 months ago
- ☆109Updated last year
- extensible collectives library in triton☆86Updated 2 months ago
- This repository contains the experimental PyTorch native float8 training UX☆224Updated 10 months ago
- ML/DL Math and Method notes☆61Updated last year
- Small scale distributed training of sequential deep learning models, built on Numpy and MPI.☆134Updated last year
- Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.☆45Updated 11 months ago
- ☆159Updated last year
- Framework to reduce autotune overhead to zero for well known deployments.☆77Updated last week