carsonpo / quadmulLinks
a fast and customizable CUDA int4 tensor core gemm
☆12Updated last year
Alternatives and similar repositories for quadmul
Users that are interested in quadmul are comparing it to the libraries listed below
Sorting:
- ☆59Updated 10 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆41Updated 7 months ago
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆45Updated 4 months ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆67Updated last year
- ☆82Updated 8 months ago
- Implement Flash Attention using Cute.☆96Updated 10 months ago
- llama INT4 cuda inference with AWQ☆55Updated 8 months ago
- [ACL 2024] A novel QAT with Self-Distillation framework to enhance ultra low-bit LLMs.☆122Updated last year
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆116Updated last year
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆144Updated last month
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆94Updated last month
- ☆56Updated 3 months ago
- Benchmark code for the "Online normalizer calculation for softmax" paper☆101Updated 7 years ago
- llm theoretical performance analysis tools and support params, flops, memory and latency analysis.☆108Updated 3 months ago
- Persistent dense gemm for Hopper in `CuTeDSL`☆15Updated 2 months ago
- ☆174Updated 2 years ago
- Fast and low-memory attention layer written in CUDA☆19Updated 2 years ago
- ☢️ TensorRT 2023复赛——基于TensorRT-LLM的Llama模型推断加速优化☆50Updated 2 years ago
- A set of examples around MegEngine☆31Updated last year
- Standalone Flash Attention v2 kernel without libtorch dependency☆111Updated last year
- The official implementation of the EMNLP 2023 paper LLM-FP4☆216Updated last year
- An easy-to-use package for implementing SmoothQuant for LLMs☆107Updated 6 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆75Updated last year
- Awesome code, projects, books, etc. related to CUDA☆24Updated 2 months ago
- [ICML 2025] Official PyTorch implementation of "FlatQuant: Flatness Matters for LLM Quantization"☆177Updated this week
- Code Repository of Evaluating Quantized Large Language Models☆132Updated last year
- ☆141Updated last year
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆223Updated 2 months ago
- ☆119Updated 2 months ago
- ☆73Updated 11 months ago