TurboQuant: Near-optimal KV cache quantization for LLM inference (3-bit keys, 2-bit values) with Triton kernels + vLLM integration

A from-scratch Prefill/Decode disaggregation inference engine for LLMs

🚀🚀 「大模型」2小时完全从0训练64M的小参数GPT！🌏 Train a 64M-parameter GPT from scratch in just 2h!

Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1

[NeurIPS2025] "AI-Researcher: Autonomous Scientific Innovation" -- A production-ready version: https://novix.science/chat

AI agents running research on single-GPU nanochat training automatically

Code repo for efficient quantized MoE inference with mixture of low-rank compensators

Official implementation of Half-Quadratic Quantization (HQQ)

FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.

A machine model for line-rate programmable switches

Leetcode for Pytorch

This repository contains the source code for P4TG, a 1 Tb/s traffic generator for Ethernet/IP networks

BSP for X-T Programmable Bare Metal Switches

Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

"Your Fully-Automated Personal AI Assistant"

Your personal AI trading assistant. Any market. Any model. Pay with USDC, not API keys.

A opensource AI trading platform in real market,

Nix packaging of the Intel Tofino SDE

barefoot platform bsp

Hands-on tutorial to learn the building blocks of the Next-Gen SDN architecture

CRS-自建Claude Code镜像，一站式开源中转服务，让 Claude、OpenAI、Gemini、Droid 订阅统一接入，支持拼车共享，更高效分摊成本，原生工具无缝使用。

UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)

Stratum is an open source silicon-independent switch operating system for software defined networks.

Ubuntu安装V2ray指南

Nano vLLM