Xianlu Bird

litellm. Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM]

dora. DORA (Dataflow-Oriented Robotic Architecture) is middleware designed to streamline and simplify the creation of AI-based robotic applications. It offers low latency, composable, and distributed dataflow capabilities. Applications are modeled as directed graphs, also referred to as pipelines.

sub2api. Sub2API 一站式开源中转服务，让 Claude、Openai 、Gemini、Grok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。

Understand-Anything. Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.

mxc. Policy-driven, layered isolation and containment

pegaflow. High-performance KV cache storage for LLM inference — GPU offloading, SSD caching, and cross-node sharing via RDMA. Works with vLLM and SGLang.

ax. Google's open source distributed agent runtime

lance. Open Lakehouse Format for Multimodal AI. Convert from Parquet in 2 lines of code for 100x faster random access, vector index, and data versioning. Compatible with Pandas, DuckDB, Polars, Pyarrow, and PyTorch with more integrations coming..

x-algorithm. Algorithm powering the For You feed on X

juicefs. JuiceFS is a distributed POSIX file system built on top of Redis and S3.

alluxio. Alluxio, data orchestration for analytics and machine learning in the cloud

agent-sandbox. agent-sandbox enables easy management of isolated, stateful, singleton workloads, ideal for use cases like AI agent runtimes.

dra-driver-nvidia-gpu. DRA Driver for NVIDIA GPUs

aitune. NVIDIA AITune is an inference toolkit designed for tuning and deploying Deep Learning models with a focus on NVIDIA GPUs.

ann-book. 《人人能懂的人工神经网络》

LMCache. LMCache: Supercharge Your LLM with the Fastest KV Cache Layer

aiperf. AIPerf is a comprehensive benchmarking tool that measures the performance of generative AI models served by your preferred inference solution.

nixl. NVIDIA Inference Xfer Library (NIXL)

superpowers. An agentic skills framework & software development methodology that works.

modelexpress. Model Express is a Rust-based component meant to be placed next to existing model inference systems to speed up their startup times and improve overall performance.

volcano. A Cloud Native Batch System (Project under CNCF)

libnvidia-container. NVIDIA container runtime library

k8s-device-plugin. NVIDIA device plugin for Kubernetes

lima. Linux virtual machines, with a focus on running containers

grove. Kubernetes enhancements for Network Topology Aware Gang Scheduling & Autoscaling

nats-server. High-Performance server for NATS.io, the cloud and edge native messaging system.

llm-d. Achieve state of the art inference performance with modern accelerators on Kubernetes

A2A. Agent2Agent (A2A) is an open protocol enabling communication and interoperability between opaque agentic applications.

OpenShell. OpenShell is the safe, private runtime for autonomous AI agents.

aicr. Tooling for optimized, validated, and reproducible GPU-accelerated AI runtime in Kubernetes

cloud-hypervisor. A Virtual Machine Monitor for modern Cloud workloads. Features include CPU, memory and device hotplug, support for running Windows and Linux guests, device offload with vhost-user and a minimal compact footprint. Written in Rust with a strong focus on security.

etcd-diagnosis. A diagnostic tool for etcd that helps identify and resolve cluster issues efficiently

dcgm-exporter. NVIDIA GPU metrics exporter for Prometheus leveraging DCGM

NVSentinel. NVSentinel is a cross-platform fault remediation service designed to rapidly remediate runtime node-level issues in GPU-accelerated computing environments

OpenMAIC. Open Multi-Agent Interactive Classroom — Get an immersive, multi-agent learning experience in just one click

Qwen-Agent. Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.

claude-agent-sdk-python. Python

Awesome-ML-SYS-Tutorial. My learning notes for ML SYS.

InfraTech. 分享AI Infra知识&代码练习：PyTorch、vLLM/SGLang、slime/vime框架入门⚡️、性能加速🚀、大模型基础🧠、AI软硬件🔧等

txt_files.

mini-nanoGPT. One-click training of your own GPT. Training a GPT has never been easier for beginners. / 一键预训练+SFT一个属于自己的LLM，0基础训练GPT原来可以这么简单？

mini-sglang. A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

ai-performance-engineering. Code, labs, and resources for O'Reilly AI Systems Performance Engineering: GPU optimization, distributed training, inference scaling, and full-stack tuning.

rbg. A workload for deploying LLM inference services on Kubernetes

lws. LeaderWorkerSet: An API for deploying a group of pods as a unit of replication

Build-A-Large-Language-Model-CN. 《Build a Large Language Model (From Scratch)》是一本深入探讨大语言模型原理与实现的电子书，适合希望深入了解 GPT 等大模型架构、训练过程及应用开发的学习者。为了让更多中文读者能够接触到这本极具价值的教材，我决定将其翻译成中文，并通过 GitHub 进行开源共享。

openagents. OpenAgents - AI Agent Networks for Open Collaboration

deer-flow. An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.

holmesgpt. SRE Agent - CNCF Sandbox Project

langgraph. Build resilient agents.

firecracker. Secure and fast microVMs for serverless computing.

mcp-go. A Go implementation of the Model Context Protocol (MCP), enabling seamless integration between LLM applications and external data sources and tools.

k8sgpt. Giving Kubernetes Superpowers to everyone

intro-llm-code. Python

nano-vllm. Nano vLLM

cursor-free-everyday. 完全免费, 自动获取新账号,一键重置新额度, 解决机器码问题, 自动满额度

clash-for-linux-install. 😼 优雅地使用基于 clash/mihomo 的代理环境

Mooncake. Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

robusta. Better Prometheus alerts for Kubernetes - smart grouping, AI enrichment, and automatic remediation

LLMs-from-scratch. Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

mcp-shell-server. Secure MCP server for whitelisted shell command execution with stdin, argv pipelines, timeouts, and structured audit logging.

ray. Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

kuberay. A toolkit to run Ray applications on Kubernetes

HAMi. Heterogeneous GPU Sharing on Kubernetes

LLMs-from-scratch-CN. LLMs-from-scratch项目中文翻译

dify. Production-ready platform for agentic workflow development.

LeetCUDA. 📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉

dynamo. A Datacenter Scale Distributed Inference Serving Framework

ai-agents-for-beginners. 12 Lessons to Get Started Building AI Agents

fluid. Fluid, elastic data abstraction and acceleration for BigData/AI applications in cloud. (Project under CNCF)

servers. Model Context Protocol Servers

llm-action. 本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）

sglang. SGLang is a high-performance serving framework for large language models and multimodal models.

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

FlashMLA. FlashMLA: Efficient Multi-head Latent Attention Kernels

Hands-On-Large-Language-Models. Official code repo for the O'Reilly Book - "Hands-On Large Language Models"

aibrix. Cost-efficient and pluggable Infrastructure components for GenAI inference

Hands-On-Large-Language-Models-CN. 中文翻译的 Hands-On-Large-Language-Models (hands-on-llms)，动手学习大模型

DeepEP. DeepEP: an efficient expert-parallel communication library

open-infra-index. Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation

DeepSeek-R1.

leedl-tutorial. 《李宏毅深度学习教程》（李宏毅老师推荐👍，苹果书🍎），PDF下载地址：https://github.com/datawhalechina/leedl-tutorial/releases

TinyZero. Minimal reproduction of DeepSeek R1-Zero

llm-cookbook. 面向开发者的 LLM 入门教程，吴恩达大模型系列课程中文版

server. The Triton Inference Server provides an optimized cloud and edge inferencing solution.

DeepLearning. Python for《Deep Learning》，该书为《深度学习》(花书) 数学推导、原理剖析与源码级别代码实现

auger. Directly access data objects stored in etcd by Kubernetes.

clusterpedia. The Encyclopedia of Kubernetes clusters

live. ✯ 可直连访问的电视/广播图标库与相关工具项目 ✯ 🔕 永久免费 直连访问 完整开源 不断完善的台标 支持IPv4/IPv6双栈访问 🔕

iptv. Collection of publicly available IPTV channels from all over the world

godel-scheduler. a unified scheduler for online and offline tasks

koordinator. A QoS-based scheduling system brings optimal layout and status to workloads such as microservices, web services, big data jobs, AI jobs, etc.

kubevirt. Kubernetes Virtualization API and runtime in order to define and manage virtual machines.

ovs. Open vSwitch

tiny-gpu. A minimal GPU design in Verilog to learn how GPUs work from the ground up

pwru. Packet, where are you? -- eBPF-based Linux kernel networking debugger

ovn-kubernetes. A robust Kubernetes networking platform

bpftop. bpftop provides a dynamic real-time view of running eBPF programs. It displays the average runtime, events per second, and estimated total CPU % for each program.

kube-ovn. A Bridge between SDN and Cloud Native (Project under CNCF)

DARC. Decentralized Autonomous Regulated Company (DARC), a company virtual machine that runs on any EVM-compatible blockchain, with on-chain law system, multi-level tokens and dividends mechanism.

slim. Slim(toolkit): Don't change anything in your container image and minify it by up to 30x (and for compiled languages even more) making it secure too! (free and open source)

kubeprober. Large-scale Kubernetes cluster diagnostic tool.

kubeeye. KubeEye aims to find various problems on Kubernetes, such as application misconfiguration, unhealthy cluster components and node problems.

nydus-snapshotter. A containerd snapshotter with data deduplication and lazy loading in P2P fashion

my-tv. 我的电视 电视直播软件，安装即可使用

accelerated-container-image. A production-ready remote container image format (overlaybd) and snapshotter based on block-device.

webkubectl. Run kubectl command in Web Browser.

deepflow. eBPF Observability - Distributed Tracing and Profiling

clash-rules. 🦄️ 🎃 👻 Clash Premium 规则集(RULE-SET)，兼容 ClashX Pro、Clash for Windows 等基于 Clash Premium 内核的客户端。

apiserver-network-proxy. Go

1024app-ios. 草榴官方客户端，小草客户端，iOS

hubble. Hubble - Network, Service & Security Observability for Kubernetes using eBPF

ebpf. ebpf-go is a pure-Go library to read, modify and load eBPF programs and attach them to various hooks in the Linux kernel.

run. 润学全球官方指定GITHUB，整理润学宗旨、纲领、理论和各类润之实例；解决为什么润，润去哪里，怎么润三大问题； 并成为新中国人的核心宗教，核心信念。

bcc. BCC - Tools for BPF-based Linux IO analysis, networking, monitoring, and more

learning-ebpf. Learning eBPF, published by O'Reilly - out now! Here's where you'll find a VM config for the examples, and more

QmaoTai. 抢茅台2023最新，京东抢茅台，淘宝抢茅台 QmaoTai

MaoTai_GUIT. 天猫 TaoBao i茅台 iMatoTai JD 京东抢购、京东抢茅台 Windows 端、开箱即用无需配置环境。开发在即（开源协议采用 Apache License）

higress. 🤖 AI Gateway | AI Native API Gateway

pixie. Instant Kubernetes-Native Application Observability

kubeskoop. Network monitoring & diagnosis suite for Kubernetes

llama. Inference code for Llama models

sysbox. An open-source, next-generation "runc" that empowers rootless containers to run workloads such as Systemd, Docker, Kubernetes, just like VMs.

whisper-ctranslate2. Whisper command line client compatible with original OpenAI client based on CTranslate2.

faster-whisper. Faster Whisper transcription with CTranslate2

whisper. Robust Speech Recognition via Large-Scale Weak Supervision

FT-seed. 富途种子 浇水:seedling:,施肥:trident:,签到:clock3:,阅读新闻:newspaper:,观看教程:tv:,导出种子包:package:,自动化脚本

kueue. Kubernetes-native Job Queueing

crane. Crane is a FinOps Platform for Cloud Resource Analytics and Economics in Kubernetes clusters. The goal is not only to help users to manage cloud cost easier but also ensure the quality of applications.

velero. Backup and migrate Kubernetes applications and their persistent volumes

copier. Copier for golang, copy value from struct to struct and more

open-gpu-kernel-modules. NVIDIA Linux open GPU kernel module source

openyurt. OpenYurt - Extending your native Kubernetes to edge(project under CNCF)

superedge. An edge-native container management system for edge computing

zadig. Zadig: An AI-powered, cloud-native, distributed DevOps platform designed for developers

openresty. High Performance Web Platform Based on Nginx and LuaJIT

karmada. Open, Multi-Cloud, Multi-Cluster Kubernetes Orchestration

gcache. An in-memory cache library for golang. It supports multiple eviction policies: LRU, LFU, ARC

tcpcopy. An online request replication and TCP stream replay tool, ideal for real testing, performance testing, stability testing, stress testing, load testing, smoke testing, and more.

gin-metrics. gin-gonic/gin metrics for prometheus.

secure. HTTP middleware for Go that facilitates some quick security wins.

fasthttp. Fast HTTP package for Go. Tuned for high performance. Zero memory allocations in hot paths. Up to 10x faster than net/http

kubediff. Kubediff: a tool for Kubernetes to show differences between running state and version controlled configuration.

overlaybd. Overlaybd: a block based remote image format. The storage backend of containerd/accelerated-container-image.

tapp. Another great app kind for Kubernetes!

kubefed. Kubernetes Cluster Federation

kudo. Kubernetes Universal Declarative Operator (KUDO)

descheduler. Descheduler for Kubernetes

chaos-mesh. A Chaos Engineering Platform for Kubernetes.

ackctl. Go

argocd-notifications. Notifications for Argo CD

controller-tools. Tools to use with the controller-runtime libraries

gitops-engine. Democratizing GitOps

helm-diff. A helm plugin that shows a diff explaining what a helm upgrade would change

cdk8s. Define Kubernetes native apps and abstractions using object-oriented programming

applicationset. The ApplicationSet controller manages multiple Argo CD Applications as a single ApplicationSet unit, supporting deployments to large numbers of clusters, deployments of large monorepos, and enabling secure Application self-service.

typescript-tutorial. TypeScript 入门教程

scheduler-plugins. Repository for out-of-tree scheduler plugins based on scheduler framework.

script. Making it easy to write shell-like scripts in Go

flagger. Progressive delivery Kubernetes operator (Canary, A/B Testing and Blue/Green deployments)

trackerslist. Updated list of public BitTorrent trackers

uilive. uilive is a go library for updating terminal output in realtime

spark-operator. Kubernetes operator for managing the lifecycle of Apache Spark applications on Kubernetes.

serverless-application-model. The AWS Serverless Application Model (AWS SAM) transform is a AWS CloudFormation macro that transforms SAM templates into CloudFormation templates.

quay. Build, Store, and Distribute your Applications and Containers

goldilocks. Get your resource requests "Just Right"

gops. A tool to list and diagnose Go processes currently running on your system

grpc-proxy. gRPC proxy is a Go reverse proxy that allows for rich routing of gRPC calls with minimum overhead.

cgroups. cgroups package for Go

tag-app-delivery. 📨🚚CNCF App Delivery TAG

siege. Siege is an http load tester and benchmarking utility

k8s-source-code-analysis. 《k8s-1.13版本源码分析》

install-docker. Scripts for docker-machine to install a particular docker version

label-studio. Label Studio is a multi-type data labeling and annotation tool with standardized output format

octant. Highly extensible platform for developers to better understand the complexity of Kubernetes clusters.

ack-ram-authenticator. Using Alibaba Cloud credentials to authenticate to a Kubernetes cluster

rudr. A Kubernetes implementation of the Open Application Model specification

docker-gc. INACTIVE: Docker garbage collection of containers and images

yunikorn-core. Apache YuniKorn Core

hashmap. A Golang lock-free thread-safe HashMap optimized for fastest read access.

node-problem-detector. This is a place for various problem detectors running on the Kubernetes nodes.

chisel. A fast TCP/UDP tunnel over HTTP

goproxy. A global proxy for Go modules.

argo-kube-notifier. Argo Kube Notifier controller monitors Kubernetes resources

d2l-zh. 《动手学深度学习》：面向中文读者、能运行、可讨论。中英文版被70多个国家的500多所大学用于教学。

kruise. Automated management of large-scale applications on Kubernetes (incubating project under CNCF)

argo-rollouts. Progressive Delivery for Kubernetes

kube-batch. A batch scheduler of kubernetes for high performance workload, e.g. AI/ML, BigData, HPC

konstellate. Free and Open Source GUI to Visualize Kubernetes Applications.

keda. KEDA is a Kubernetes-based Event Driven Autoscaling component. It provides event driven scale for any container running in Kubernetes

tensorflow2_tutorials_chinese. tensorflow2中文教程，持续更新(当前版本:tensorflow2.0)，tag: tensorflow 2.0 tutorials

sharness. Shell library to test your tools like Git does

argo-helm. ArgoProj Helm Charts

kustomize. Customization of kubernetes YAML configurations

flink-training-course. Flink 中文视频课程（持续更新...）

springboot-usage.

nndl. 邱锡鹏《神经网络与深度学习》（蒲公英书）理论书 v2 与通识版

failpoint. An implementation of failpoints for Golang.

gomodules.

Deep-Learning-21-Examples. 《21个项目玩转深度学习———基于TensorFlow的实践详解》配套代码