tlc-pack / cutlass_fpA_intB_gemm
A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer
☆82Updated 6 months ago
Related projects: ⓘ
- Standalone Flash Attention v2 kernel without libtorch dependency☆93Updated last week
- play gemm with tvm☆81Updated last year
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆46Updated last month
- TiledCUDA is a highly efficient kernel template library designed to elevate CUDA C’s level of abstraction for processing tiles.☆114Updated last week
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆20Updated last week
- ☆67Updated last week
- llama INT4 cuda inference with AWQ☆46Updated 2 months ago
- ☆23Updated this week
- ☆34Updated 2 years ago
- ☆77Updated last year
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆81Updated 2 months ago
- ☆133Updated 2 months ago
- ☆32Updated 3 months ago
- DietCode Code Release☆59Updated 2 years ago
- An extention of TVMScript to write simple and high performance GPU kernels with tensorcore.☆49Updated last month
- PyTorch bindings for CUTLASS grouped GEMM.☆41Updated 3 weeks ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆40Updated last week
- ☆56Updated this week
- ☆18Updated 5 months ago
- ☆70Updated 6 months ago
- ☆52Updated this week
- GPTQ inference TVM kernel☆35Updated 4 months ago
- ☆138Updated 2 months ago
- Benchmark code for the "Online normalizer calculation for softmax" paper☆52Updated 6 years ago
- ☆140Updated 4 months ago
- SparseTIR: Sparse Tensor Compiler for Deep Learning☆129Updated last year
- Benchmark scripts for TVM☆73Updated 2 years ago
- This is a demo how to write a high performance convolution run on apple silicon☆52Updated 2 years ago
- ☆14Updated 2 years ago
- ☆102Updated 3 months ago