ssiu / flash-attention-turingLinks
☆70Updated this week
Alternatives and similar repositories for flash-attention-turing
Users that are interested in flash-attention-turing are comparing it to the libraries listed below
Sorting:
- Fast and memory-efficient exact attention☆214Updated last week
- a simple Flash Attention v2 implementation with ROCM (RDNA3 GPU, roc wmma), mainly used for stable diffusion(ComfyUI) in Windows ZLUDA en…☆51Updated last year
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆276Updated 6 months ago
- A high-throughput and memory-efficient inference and serving engine for LLMs☆114Updated this week
- A safetensors extension to efficiently store sparse quantized tensors on disk☆244Updated this week
- Development repository for the Triton language and compiler☆140Updated this week
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,011Updated last year
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆154Updated 5 months ago
- ☆206Updated 9 months ago
- Model Compression Toolbox for Large Language Models and Diffusion Models☆753Updated 5 months ago
- Ahead of Time (AOT) Triton Math Library☆88Updated this week
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆751Updated 6 months ago
- DFloat11 [NeurIPS '25]: Lossless Compression of LLMs and DiTs for Efficient GPU Inference☆603Updated 2 months ago
- An easy-to-use package for implementing SmoothQuant for LLMs☆110Updated 10 months ago
- ☆172Updated this week
- AI Tensor Engine for ROCm☆351Updated this week
- DashInfer is a native LLM inference engine aiming to deliver industry-leading performance atop various hardware architectures, including …☆274Updated 6 months ago
- Fast low-bit matmul kernels in Triton☆429Updated last week
- CPM.cu is a lightweight, high-performance CUDA implementation for LLMs, optimized for end-device inference and featuring cutting-edge tec…☆226Updated 3 weeks ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆250Updated last week
- 8-bit CUDA functions for PyTorch☆70Updated 4 months ago
- NVIDIA Linux open GPU with P2P support☆129Updated this week
- 🎯An accuracy-first, highly efficient quantization toolkit for LLMs, designed to minimize quality degradation across Weight-Only Quantiza…☆845Updated this week
- ☆163Updated 7 months ago
- llama INT4 cuda inference with AWQ☆54Updated last year
- GPTQ inference Triton kernel☆321Updated 2 years ago
- ☆105Updated last year
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆44Updated 11 months ago
- AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (N…☆12Updated last year
- ☆130Updated last year