Starred repositories
A modular, parametrizable, and highly flexible Data Movement Accelerator (DMA)
A Fast, Low-Overhead On-chip Network
A Chisel RTL generator for network-on-chip interconnects
HBFSim (HBF Sim): an open-source High-Bandwidth Flash (HBF) simulator. Runs real LLM inference on a real GPU and injects modeled HBF timing, capacity and thermal effects while the workload executes.
LLMServingSim 2.0: A Unified Simulator for Heterogeneous and Disaggregated LLM Serving Infrastructure
Repository for MLCommons Chakra schema and tools
Official inference framework for 1-bit LLMs
KVarN is a native vLLM KV-cache quantization backend for your agents: 3-5x more context, throughput above FP16, and FP16-level accuracy. Calibration-free, one flag.
Code for the ProteinMPNN paper
User friendly and accurate binder design pipeline
Easy to use suite for designing a wide range of protein binder modalities.
Learning in infinite dimension with neural operators.
UniSparse: An Intermediate Language for General Sparse Format Customization (OOPSLA'24)
MCP-based Computer Use service for Qwen Code and any AI agent — controls macOS, Linux, and Windows via accessibility APIs.
👾 Open Computer Use – Open-Source Alternative to Codex Computer Use
Zephyr OS will become the OSAL (OS Abstraction Layer) for CAR-OS (NammaAUTOSAR)
ISO 14229 UDS diagnostics stack for Zephyr and FreeRTOS. YAML-configured. ASIL-B-oriented architecture. DoIP + CAN.
Lightweight coding agent that runs in your terminal
AI 基础知识 - GPU 架构、CUDA 编程、大模型基础及AI Agent 相关知识。
DeepSeek Harness: Everything is a Plugin.