Let your Claude able to think

A SOTA quantization algorithm for high-accuracy low-bit LLM inference, seamlessly optimized for CPU/XPU/CUDA, with multi-datatype support and full compatibility with vLLM, SGLang, and Transformers.

Intel® NPU Acceleration Library

An innovative library for efficient LLM inference via low-bit quantization

how to optimize some algorithm in cuda.

⚡ Build your chatbot within minutes on your favorite device; offer SOTA compression techniques for LLMs; run LLMs efficiently on Intel Platforms⚡

SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime

Writing a minimal x86-64 JIT compiler in C++

🤗 Optimum Intel: Accelerate inference with Intel optimization tools

Intel® Extension for TensorFlow*

MLIR Sample dialect

MLIRX is now defunct. Please see PolyBlocks - https://docs.polymagelabs.com

Parallel Algorithm Scheduling Library

This is an implementation of sgemm_kernel on L1d cache.

an educational compiler intermediate representation

A primitive library for neural network

A list of awesome compiler projects and papers for tensor computation and deep learning.

Fast sparse deep learning on CPUs

Assembler for NVIDIA Maxwell architecture

Samples for Intel® oneAPI Toolkits

LLVM Optimization to extract a function, embedded in its intermediate representation in the binary, and execute it using the LLVM Just-In-Time compiler.

Transform ONNX model to PyTorch representation

Intel Data Parallel C++ (and SYCL 2020) Tutorial.

LightSeq: A High Performance Library for Sequence Processing and Generation

Representation and Reference Lowering of ONNX Models in MLIR Compiler Infrastructure

ipex verbose toolkit

Python Framework for sparse neural networks

a c++/cuda template library for tensor lazy evaluation