flagos-ai / FlagAttentionLinks
A collection of memory efficient attention operators implemented in the Triton language.
☆287Updated last year
Alternatives and similar repositories for FlagAttention
Users that are interested in FlagAttention are comparing it to the libraries listed below
Sorting:
- PyTorch bindings for CUTLASS grouped GEMM.☆174Updated last week
- ☆103Updated last year
- PyTorch bindings for CUTLASS grouped GEMM.☆135Updated 6 months ago
- ☆154Updated 9 months ago
- ☆253Updated last year
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆301Updated this week
- Zero Bubble Pipeline Parallelism☆442Updated 7 months ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆276Updated 5 months ago
- flash attention tutorial written in python, triton, cuda, cutlass☆462Updated 7 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆242Updated last month
- Applied AI experiments and examples for PyTorch☆311Updated 4 months ago
- This repository contains integer operators on GPUs for PyTorch.☆223Updated 2 years ago
- A Easy-to-understand TensorOp Matmul Tutorial☆397Updated 2 months ago
- FlagGems is an operator library for large language models implemented in the Triton Language.☆803Updated this week
- Dynamic Memory Management for Serving LLMs without PagedAttention☆448Updated 6 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆43Updated 9 months ago
- An easy-to-use package for implementing SmoothQuant for LLMs☆110Updated 8 months ago
- Utility scripts for PyTorch (e.g. Make Perfetto show some disappearing kernels, Memory profiler that understands more low-level allocatio…☆72Updated 3 months ago
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆331Updated last year
- LLM training technologies developed by kwai☆67Updated last month
- ☆97Updated 8 months ago
- Accelerating MoE with IO and Tile-aware Optimizations☆351Updated last week
- ☆112Updated 7 months ago
- Pipeline Parallelism Emulation and Visualization☆74Updated 6 months ago
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆151Updated 4 months ago
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆230Updated 2 years ago
- Allow torch tensor memory to be released and resumed later☆191Updated 3 weeks ago
- Cataloging released Triton kernels.☆278Updated 3 months ago
- nnScaler: Compiling DNN models for Parallel Training☆121Updated 3 months ago
- ☆141Updated last year