ankan-ban / llama_cu_awqLinks
llama INT4 cuda inference with AWQ
☆54Updated 6 months ago
Alternatives and similar repositories for llama_cu_awq
Users that are interested in llama_cu_awq are comparing it to the libraries listed below
Sorting:
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆94Updated 3 weeks ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆113Updated last year
- Standalone Flash Attention v2 kernel without libtorch dependency☆111Updated 10 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆74Updated 11 months ago
- Benchmark code for the "Online normalizer calculation for softmax" paper☆96Updated 7 years ago
- ☆96Updated 11 months ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆63Updated 11 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆39Updated 5 months ago
- play gemm with tvm☆91Updated 2 years ago
- ☆80Updated 6 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆93Updated 2 months ago
- SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs☆51Updated 4 months ago
- ☆91Updated 2 months ago
- ☆37Updated last year
- ☆60Updated 3 months ago
- ☆150Updated last year
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆93Updated last month
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆183Updated 6 months ago
- DeeperGEMM: crazy optimized version☆71Updated 3 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆107Updated 2 months ago
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆216Updated last year
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆40Updated last month
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆56Updated this week
- SparseTIR: Sparse Tensor Compiler for Deep Learning☆137Updated 2 years ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆260Updated 3 weeks ago
- This repository contains integer operators on GPUs for PyTorch.☆211Updated last year
- ☆85Updated 9 months ago
- Implement Flash Attention using Cute.☆92Updated 7 months ago
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆137Updated 4 months ago
- CUDA 8-bit Tensor Core Matrix Multiplication based on m16n16k16 WMMA API☆31Updated last year