Universal LLM Deployment Engine with ML Compilation

Code Repository of Evaluating Quantized Large Language Models

Implementation for MatMul-free LM.

The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.

Transformer related optimization, including BERT, GPT

TigerBot: A multi-language multi-task LLM

a lightweight LLM model inference framework

Aligning pretrained language models with instruction data generated by themselves.

[ICML'21 Oral] I-BERT: Integer-only BERT Quantization

A retargetable MLIR-based machine learning compiler and runtime toolkit.

CUDA Templates for Linear Algebra Subroutines

A simple tool to profile performance of multiple combinations of GEMM of cuBLAS

flexible-gemm conv of deepcore

Compiler for Neural Network hardware accelerators

A domain specific language to express machine learning workloads.

A CPU tool for benchmarking the peak of floating points

Subpart source code of of deepcore v0.7

Efficient GPU kernels for block-sparse matrix multiplication and convolution

OpenBLAS is an optimized BLAS library based on GotoBLAS2 1.13 BSD version.

Library for specialized dense and sparse matrix operations, and deep learning primitives.

Winograd minimal convolution algorithm generator for convolutional neural networks.

Library for fast image convolution in neural networks on Intel Architecture

a heterogeneous multiGPU level-3 BLAS library

Assembler for NVIDIA Maxwell architecture

An MPI-based C++ or Python library for easy distributed pipeline processing

Boost.org mpi module

CudaPAD is a PTX/SASS viewer for NVIDIA Cuda kernels and provides an on-the-fly view of the assembly.