an agent-native sandbox runtime

Performance tests and benchmarks

Dify is an open-source LLM app development platform. Dify's intuitive interface combines AI workflow, RAG pipeline, agent capabilities, model management, observability features and more, letting yo…

Production-Grade Container Scheduling and Management

An Open-source, self-hosted AI model hub with Hugging Face compatibility, accelerating vLLM/SGLang performance.

RDMA core userspace libraries and daemons

Prometheus exporter for collection of RDMA (RoCE) NIC statistics from Linux hosts

Nano vLLM

Agent Substrate: the core system

Instant, Concurrent, Secure & Lightweight Sandbox for AI Agents.

Secure and fast microVMs for serverless computing.

NVIDIA GPU Operator creates, configures, and manages GPUs in Kubernetes

AI 基础知识 - GPU 架构、CUDA 编程、大模型基础及AI Agent 相关知识。

Dynamic sub-dir volume provisioner on a remote NFS server.

Resource Topology exporter for Topology Aware Scheduler

ONCache: A Cache-Based Low-Overhead Container Overlay Network

A hacker's userspace TCP/IP stack

dperf: High-Performance Network Load Testing Tool Based on DPDK

Rust Kubernetes client and controller runtime

Tools to use with the controller-runtime libraries

A fast and lightweight fully featured OCI runtime and C library for running containers

transfer confluence pages to markdown style

An effort to comprehensively document eBPF

Layer-4 eBPF load balancer on the Kubernetes, with strengthened kube-proxy replacement and multi-clusters connectivity

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without …

Lightweight service mesh for Kubernetes East-West and North-South traffic management, uses eBPF for layer4 and pipy proxy for layer7 traffic management, support multi cluster network.