Project AIRI is the most popular open-source AI companion platform — a self-hosted recreation of Neuro-sama that brings AI-powered virtual characters into your world across web, desktop, and mobile. The system renders Live2D, Spine, and VRM 3D character models with auto-blink, eye tracking, and lip-sync driven by real-time voice synthesis, while the xsAI abstraction layer connects to 40+ LLM providers including OpenAI GPT-4, Anthropic Claude, Google Gemini, DeepSeek, and local models via Ollama and OpenRouter. Built from day one on WebGPU, WebAudio, Web Workers, WebAssembly, and WebSocket technologies, the browser version runs entirely client-side with PWA offline support while the server runtime enables persistent memory via PostgreSQL with pgvector embeddings and DuckDB WASM for client-side storage. The Minecraft agent plays autonomously using mineflayer with pathfinding, and a Factorio integration provides cooperative gameplay. Social integrations deploy your companion as a Discord bot joining voice channels, a Telegram bot, and a Twitter/X agent posting and replying autonomously. The desktop Stage Tamagotchi app provides an always-on-screen companion for Windows and macOS, while Stage Pocket brings the experience to mobile. Voice features include client-side speech recognition via VAD, multiple TTS providers including ElevenLabs, and screen vision capabilities. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Born at Brazilian media giant Globo.com, Thumbor answers imaging CDNs like Imgix and Cloudinary with an HTTP service where every image variant is just a URL. Ask for /300x200/smart/your-image.jpg and Thumbor fetches the original, crops and resizes on demand, and caches the result - one source file, unlimited renditions, no batch pre-generation pipeline. The "smart" in the URL is the signature feature: OpenCV-based face detection finds people in the frame and crops around them (no more thumbnails with severed heads), and when no faces exist, feature detection finds visually important corners and computes a weighted center of mass as the focal point. Beyond cropping, a chainable filter pipeline handles brightness, contrast, grayscale, blur, red-eye removal, rounded corners, rotation, watermarks, and format conversion with quality control - applied in order via URL segments. All common image formats work out of the box, and every layer is pluggable: loaders (HTTP, local, S3), storages and result storages (local, S3, Ceph, and community backends), engines, optimizers, filters, and even custom detectors, with the awesome-thumbor list cataloging the ecosystem. URL signing prevents abuse of your processing capacity. Integrations exist for Django, Rails, Node, WordPress, and most frameworks. MIT-licensed, battle- tested for over a decade.

Cockpit puts Linux server administration in a web browser, rendering storage, networking, containers, and virtual machines as visual panels you click through instead of command-line syntax you memorize. It ships in every major Linux distribution and activates through systemd socket activation, consuming zero resources until someone connects on port 9090. What makes it different from other admin panels: every change through the web interface appears instantly at the command line, and every terminal change reflects in the browser in real time. No abstraction layer sits between the UI and the operating system, which means no configuration drift. Storage management covers disk partitioning, RAID arrays, LUKS encryption, LVM volumes, and NFS shares. Network configuration handles bonds, bridges, VLANs, and firewall zones through visual editors. Native Podman integration pulls images, creates containers, manages pods, and provides terminal access into running processes. KVM/libvirt support handles VM creation, cloning, snapshots, and live migration with console access. A built-in terminal provides shell access when the web interface is not enough, and the journal viewer filters systemd logs by severity and unit. Multi-server management connects additional hosts over SSH, presenting a unified dashboard across your entire infrastructure from one browser tab. Runs on a dedicated RepoCloud VPS with full root SSH access and browser serial console. LGPL licensed.

MediaMTX sits between cameras and viewers as a self-hosted live media server, taking in video and audio streams over one protocol and handing them back out over another, so one IP camera feed can end up playing in a browser, a phone app, and a recording archive simultaneously. Publish with FFmpeg, GStreamer, OBS Studio, a browser, or a Raspberry Pi camera over RTSP, RTMP, SRT, WebRTC, Media-over-QUIC, HLS, MPEG-TS, or RTP, and read it back on any other protocol without a transcoding pipeline. Run dozens of camera feeds side by side, each on its own path, and keep every path available even while its publisher is offline. Record everything to disk in fMP4 or MPEG-TS as it streams, then play recordings back later through the same server instead of shipping files around. Authenticate publishers and viewers with built-in users, an external HTTP check, or JWT, and use the control API plus Prometheus metrics to wire the server into the rest of your monitoring stack. Forward streams to other servers, proxy cameras behind NAT, and fire hook commands whenever a client connects, publishes, or disconnects, so a doorbell camera can trigger a snapshot or an alert as it starts streaming. Reload the configuration while clients stay connected, and tune everything from the same mediamtx.yml file or environment variables, with no database or runtime dependencies to babysit. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Benchmarked at one million concurrent WebSocket connections and 30 million delivered messages per minute on a single server, Centrifugo is a self-hosted real-time messaging server that integrates with any backend language — Node.js, Python, Go, Ruby, PHP, Java — without requiring architectural changes to your existing application. It handles persistent connections over WebSocket, HTTP-streaming, Server-Sent Events, gRPC, and experimental WebTransport, routing messages through a channel-based PUB/SUB model with automatic reconnection recovery that replays missed messages from hot channel history. Built-in horizontal scaling operates through Redis (including Redis Cluster, AWS ElastiCache, Valkey, KeyDB, and DragonflyDB), NATS, or PostgreSQL brokers, allowing multi-node deployments without external message queues. The server provides JWT-based authentication using HMAC or RSA keys, granular channel permissions, online presence tracking showing who is currently subscribed to a channel, and an RPC proxy that forwards client calls to your backend over HTTP or gRPC. The embedded ReactJS admin web interface displays real-time connection statistics, node information, and provides an interactive API console for publishing messages, managing subscriptions, and inspecting channel history. Official client SDKs are available for JavaScript, Dart, Go, Swift, and Java, with unidirectional transports requiring no SDK at all. Centrifugo also supports delta updates to minimize bandwidth, connection event proxying, and Prometheus-compatible metrics for Grafana monitoring. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Images, text, audio, video, HTML, PDFs, and time series, labeled in one tool with a standardized output format: Label Studio is the open-source data labeling platform for building training datasets. Computer vision tasks cover classification, object detection (boxes, polygons, ellipses, keypoints), and semantic segmentation; audio work spans transcription, speaker diarization, and emotion recognition; NLP handles named entity recognition and document classification with taxonomies up to 10,000 classes; and GenAI workflows support LLM fine-tuning data and RLHF response ranking. Labeling interfaces are fully configurable with an XML-like templating language, so the UI matches the task instead of the reverse. The ML backend SDK turns any model into a connected web server for pre-annotation (model predicts, humans verify), interactive labeling (real-time predictions as annotators draw regions or highlight text), and model evaluation - cutting annotation time dramatically on large datasets. Data imports from S3, GCS, or file uploads; the Data Manager filters and explores tasks; exports convert to the format your ML library expects via label-studio-converter. Multi-user accounts tie every annotation to its author, and webhooks, a Python SDK, and REST API embed labeling into any pipeline. Self-hosting keeps proprietary training data - often a company's most sensitive asset - entirely on your infrastructure.

The most-starred open-source project management platform on GitHub with over 55,000 stars, Plane delivers what Jira, Linear, Monday, and ClickUp charge thousands per year for — issue tracking, sprint planning, documentation, and AI-powered workflows in one unified workspace that you own and control entirely. Work items feature a rich text editor with file uploads, sub-properties, custom states, priorities, labels, assignees, and cross-referencing, organized across five customizable layout views (list, board, table, spreadsheet, Gantt) with Command-K navigation for instant access to anything. Time-boxed Cycles provide sprint planning with automatic burn-down charts, velocity tracking, and scope change detection, while Modules break complex projects into manageable deliverables with progress aggregation. Built-in Pages combine AI-powered documentation with rich formatting, image embedding, and one-click conversion of notes into actionable work items. The AI layer reads across every project, cycle, document, and thread in the workspace — agents take real assignments, triage incoming requests, assign owners, track blockers, and ship status updates automatically. Native integrations connect GitHub, GitLab, Slack, Sentry, Figma, and 50+ tools with bidirectional issue sync and PR tracking, while import pipelines migrate entire workspaces from Jira, Linear, Asana, ClickUp, or Monday in minutes. The REST API with OAuth 2.0, HMAC-signed webhooks, typed SDKs in Node.js and Python, and a native MCP server enable custom automations. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Memos, the lightweight open-source note service from the usememos project, packaged as a containerized deployment for multi-architecture Docker hosts (x86-64 and arm64): that is Usermemos. The model is frictionless capture: no folders or titles, just a chronological stream of Markdown notes with code blocks, task lists, tables, and file attachments, organized by #hashtags pulled automatically from the text. Per-memo visibility - private, protected for logged-in users, or public - lets a single instance serve as a personal journal, a shared team log, or a public microblog simultaneously. Multi-user support with authentication makes it workable for small teams, and full REST and gRPC APIs open capture and retrieval to CLIs, bots, and automation tools. The runtime is a single Go binary with a React frontend that idles around 50 MB of memory and stores content as plain Markdown in SQLite by default, with MySQL and PostgreSQL available for heavier deployments. Configuration happens through environment variables, access works over HTTP or HTTPS behind a reverse proxy, and there is no telemetry - notes stay on your server in a portable format.

Open the page, write a Markdown note, move on - Memos is a lightweight, self-hosted service built for quick capture. Instead of folders, notebooks, and titles, it presents a timeline: open the page, write a Markdown note, and move on. Notes support headings, code blocks with syntax highlighting, task lists, tables, and file attachments, with tags auto-extracted from #hashtags in the text. Each memo carries a visibility level, private, protected (logged-in users), or public, so one instance works as a personal log, a small team wiki, or a lightweight microblog. The backend is a single Go binary with a React frontend, around 50 MB of memory at runtime and a ~20 MB Docker image, so it fits comfortably on the smallest instance size with near-zero maintenance. SQLite is the default store, with MySQL and PostgreSQL supported for multi-user deployments needing more concurrency, and full REST and gRPC APIs - Connect RPC for browsers, gRPC-Gateway for external tools - make capture scriptable from CLIs, bots, and automation platforms. Fast full-text search spans all memos, pinned notes keep references handy, and a masonry view suits visual browsing. MIT-licensed with zero telemetry; content is stored as plain Markdown in a database you control, so notes remain readable, exportable, and free of proprietary formats.

MeterSphere is the open-source continuous testing platform that brings test management, API testing, and AI-powered automation into a single self-hosted environment. The Spring Boot Java backend handles test execution with the JMeter engine while the Vue.js frontend delivers a responsive interface for managing test cases, plans, defects, and reports across projects. The built-in AI assistant leverages large language models to auto-generate functional test cases and API interface definitions, reducing manual test creation effort. Test management covers the complete lifecycle from writing and reviewing cases in list or mind-map views, through test plan creation with single plans and plan groups, to defect tracking with customizable templates and workflow rules. API testing combines Postman-like ease of use with JMeter-level flexibility, supporting interface debugging with server-side and local execution, API definition with visual request and response editors, interface mocking with configurable headers and body parameters, scenario automation with visual orchestration, and detailed test reports with automatic generation. The system-organization-project hierarchy supports up to 30 users in the community edition with role-based access control, file management, and configurable notification channels. MySQL stores application data, Kafka handles message queuing, MinIO provides S3-compatible object storage, and Redis manages caching. The plugin marketplace extends testing capabilities and enables DevOps pipeline integration. On RepoCloud, deploy MeterSphere on a dedicated VPS with Docker, root SSH access, and complete control over your testing infrastructure, all under the GPLv3 license.

Reaching 45,000 GitHub stars within seven months of launch, Multica is the fastest-growing open-source platform for managing AI coding agents as first-class teammates — assign an issue to Claude Code, Codex, Cursor, Copilot, Kimi, or any of 21 supported agent CLIs and it picks up the work, comments progress in real time via WebSocket, raises blockers, and hands the result back for human review before anything merges. The Go backend (Chi router, sqlc-generated type-safe queries, gorilla/websocket) connects to PostgreSQL 17 with pgvector for semantic search across workspace history, while the Next.js 16 App Router frontend delivers workspace dashboards showing per-agent token spend, execution time, daily cost charts, and runtime status across unlimited connected machines. Agent Skills provide reusable methods, reference material, and supporting files that compound across runs — a persistent knowledge layer that makes each subsequent task faster and more accurate. Squads let a leader agent select the right specialist for subtasks, creating multi-agent workflows without manual orchestration. Review gates ensure no AI-generated code ships to main without explicit human approval. Self-host via Docker Compose or Kubernetes with full Git integration across GitHub, GitLab, Gitea, and Forgejo including self-hosted instances. The CLI and REST API make every surface scriptable, and Autopilot automations trigger agent runs from events. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed with additional conditions.

Over 400 technologies and thousands of smart devices from any manufacturer, unified under one roof: openHAB is the vendor-neutral home automation platform with a pluggable binding architecture. Each binding translates a device or service into openHAB's clean abstraction: Things expose Channels, Channels link to Items, and Items feed a rules engine that runs your home. That engine meets you at your skill level: Blockly gives non-programmers drag-and-drop visual logic, JS Scripting (GraalJS with the openhab-js library) is the modern text-based standard, the classic Rules DSL remains supported, and JSR223 opens the door to Python, Ruby, and Groovy. Time- and event-based triggers, scripts, notifications, and voice control compose into automations of any complexity, and users report decade-old rule sets still running rock solid. The Main UI handles configuration, semantic modeling, and now built-in charting - no external Grafana required. Built in Java on Apache Karaf's OSGi runtime and stewarded by the non-profit openHAB Foundation, it requires no cloud to function: everything runs locally, talking directly to your devices. Optional connectors bridge to Alexa, Google Assistant, and HomeKit, with iOS, Android, and web apps for control from anywhere.

With over 33,000 GitHub stars and a codebase written in 92.9% Rust for maximum performance and memory safety, Tabby is the most widely adopted self-hosted alternative to GitHub Copilot — delivering real-time code completions entirely on your own infrastructure with zero code leaving your network. Deploy a single Docker container on any NVIDIA CUDA, Apple Silicon Metal, AMD ROCm, or CPU-only server and connect VS Code, JetBrains IDEs (IntelliJ, PyCharm, WebStorm, GoLand), Vim, Neovim, and Emacs through native extensions. The completion engine supports a curated registry of models including StarCoder2 (1B to 15B parameters), DeepSeek-Coder, CodeLlama, CodeGemma, Qwen2.5-Coder, and Mistral Code — swappable at runtime through the admin dashboard without redeployment. Repository indexing parses your Git repositories and feeds project-specific types, function signatures, and patterns into completion context via RAG, producing suggestions that understand your codebase rather than generic boilerplate. The Answer Engine provides instant responses to code queries within the IDE, while inline chat enables contextual code editing and explanation without switching windows. The admin dashboard manages per-developer API tokens, usage analytics, and model configuration. Enterprise features include SSO via LDAP, OAuth, and SAML, role-based access control, and audit logging for compliance environments. A single RTX 4090 workstation serves a team of 10-15 developers with sub-500ms completion latency. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Up to 280 search services - Google, Bing, DuckDuckGo, Brave, Qwant, Startpage - aggregated without tracking or profiling: SearXNG is a privacy-respecting metasearch engine (AGPL-3.0, successor to Searx). Your instance queries the upstream engines on your behalf: your IP address, cookies, and search history never reach them, tracker parameters are stripped from result URLs, and an optional image proxy fetches thumbnails server-side so result pages leak nothing. It can even route outbound queries through Tor for full anonymity. Search is organized into categories - general, images, videos, news, maps, music, IT, science, files - with bang shortcuts for targeting specific engines, and every source can be enabled, disabled, or weighted per category in settings.yml. A plugin system adds calculators, hash tools, tracker removal, and unit conversions inline, and preferences (themes, safe search, languages, engine selection) persist in cookies rather than server-side accounts. The real argument for running your own instance rather than trusting a public one is control: you decide the logging policy (none), the engine mix, rate limiting, and who gets access - making it the default search backend for browsers, families, and teams that want Google-quality results without the profile.

Backed by 27,700+ GitHub stars and actively developed with the v0.9.x architectural overhaul introducing an event-driven plugin ecosystem, ArchiveBox delivers the open-source self-hosted web archiving platform that preserves websites, bookmarks, social posts, media, source code, and research material in durable formats before link rot, platform shutdowns, and censorship erase them permanently. Feed it URLs one at a time or schedule automated imports from browser history, RSS feeds, Pocket, Pinboard, Instapaper, Wallabag, Shaarli, JSON, CSV, HTML bookmark exports, and browser extension captures. Each snapshot stores redundant copies as original HTML, rendered single-file HTML via SingleFile, full-page PDF, screenshot PNG, WARC archive, article text via Readability, favicons, HTTP headers, media files via yt-dlp, git repository clones, and structured metadata in SQLite. The Docker Compose deployment bundles Chrome, wget, curl, yt-dlp, SingleFile, and Readability parsers with automatic dependency management and isolation. Interact through the self-hosted web UI with search, tagging, and admin controls, the comprehensive CLI for batch operations, the REST API for programmatic access, or the Python API for custom integrations. The new abx-plugins system enables community extractors with per-plugin configuration, while the append-only-log architecture provides resumable crawls and audit trails. S3, B2, and Google Cloud storage backends sync archives to remote storage via rclone. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

The most feature-complete open-source alternative to Google's NotebookLM — a self-hosted research platform where you upload PDFs, videos, audio files, and web pages into organized notebooks, then chat with your content, generate multi-speaker podcasts, and run semantic search across everything without sending a single byte to Google's servers. The podcast engine supports 1-4 fully customizable speakers with backstories, personalities, and expertise profiles, generating professional audio dialogue through OpenAI, ElevenLabs, Google TTS, or completely local text-to-speech via Kokoro for maximum privacy. Content processing uses token-based chunking with RAG-powered retrieval grounded in your uploaded sources, while both full-text keyword search and semantic vector search via SurrealDB enable conceptual discovery across all notebooks. The 18+ supported AI providers include OpenAI, Anthropic, Google Gemini, Groq, Ollama, LM Studio, and more — configurable per task so you can route cheap models to summarization and powerful models to analysis. Content transformations extract insights, generate summaries, create study guides, and produce structured outputs from any source material. The MCP integration connects Open Notebook to Claude Desktop, VS Code, and other MCP clients for seamless workflow integration. A full REST API on port 5055 enables complete automation of notebook management, source upload, and podcast generation. Deploy via Docker Compose with the application container, SurrealDB v2 on RocksDB, and optional TTS containers. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Point Emby at your movie, music, and photo libraries and it becomes a private streaming service: metadata and artwork arrive from TMDB and TVDB, everything lands in a polished browsable interface, and media transcodes on the fly whenever a client can't play the original format. The client reach is the selling point: native apps span Android TV, Apple TV, Google TV, Fire TV, Roku, LG and Samsung smart TVs, iOS, Android, Apple CarPlay, Android Auto, plus web browsers and desktop apps for Windows, macOS, and Linux, and DLNA devices are auto-detected for casting and remote control. Multi-user support gives each household member their own account, watch history, favorites, and recommendations, with genuinely capable parental controls: content restrictions, access schedules, time limits, and live monitoring with remote control of kids' sessions. Live TV works with hardware tuners like HDHomeRun or M3U playlists, with free guide data in the US, Canada, and UK. The server and core features are free; an optional Premiere key adds hardware-accelerated transcoding, DVR recording, offline sync, and Cinema Mode intros.

A self-hosted "second brain": Khoj indexes your own files and answers questions from them, parsing Markdown (whole Obsidian vaults included), org-mode, PDF, Word, plain text, Notion pages, GitHub repositories, and images described by a vision model, then embedding everything with sentence-transformers into a vector index for semantic search and RAG with cited sources. Any LLM backend works: local models like Llama, Qwen, or Mistral via Ollama, or cloud models like GPT, Claude, and Gemini. You can build custom agents, each with its own persona, scoped knowledge base, chat model, and tools such as web search and code execution. Scheduled automations run recurring research and deliver newsletters or notifications to your inbox, and research mode performs multi-hop web searches with inline citations. Access it from a browser, the Obsidian plugin, Emacs, desktop, or WhatsApp - all clients connect to the same self-hosted instance, making Khoj one of the few AI assistants Emacs users can point at decades of org files. Semantic search means recall works without exact keywords: "that paper about forecasting with transformers" surfaces the right PDF even when you cannot remember its title. Switching LLM backends never requires re-indexing your documents, and with a local model via Ollama, even inference stays on hardware you control - journals, research, and private notes are never sent anywhere. Python/FastAPI stack, AGPL-licensed, with PostgreSQL storage.