A high-throughput and memory-efficient inference and serving engine for LLMs

ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)

Artifact for paper "PIM is All You Need: A CXL-Enabled GPU-Free System for LLM Inference", ASPLOS 2025

GlazeWM is a tiling window manager for Windows inspired by i3wm.

[HPCA 2023] ViTCoD: Vision Transformer Acceleration via Dedicated Algorithm and Accelerator Co-Design

A co-design architecture on sparse attention

AMBA bus generator including AXI, AHB, and APB

Fast and memory-efficient exact attention

SiT Dataset: Socially Interactive Pedestrian Trajectory Dataset for Social Navigation Robots [NeurIPS 2023]

Verilog AXI components for FPGA implementation

IC implementation of TPU

Fast and accurate DRAM power and energy estimation tool

A Fast and Extensible DRAM Simulator, with built-in support for modeling many different DRAM technologies including DDRx, LPDDRx, GDDRx, WIOx, HBMx, and various academic proposals. Described in the…

The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.

A FPGA supported RISC-V CPU with 5-stage pipeline implemented in Verilog HDL

Inference code for AI Challenge (Dec 2020)

TernGEMM: General Matrix Multiply Library with Ternary Weights for Fast DNN Inference

Layer-wise Pruning of Transformer Heads for Efficient Language Modeling

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )

YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite