The catalog.

High-resolution text/image to 3D by Tencent (2025) — generates textured meshes with photorealistic detail. Two-stage pipeline: shape gen + texture gen. Quality on par with closed-source commercial 3D AI. NVIDIA GPU 24GB+ VRAM.

Single image → 3D mesh in 0.5 seconds. Joint research by Stability AI + Tripo AI. State-of-art fast 3D reconstruction. Output: OBJ/GLB with vertex colors. NVIDIA GPU 6GB+ VRAM.

Autonomous AI agent framework with web UI. Self-improving, tool-calling, multi-modal. Code execution sandboxed via Docker.

Full-stack RAG knowledge base with built-in chat UI. Workspaces, document ingestion, multi-LLM, embedded vector DB.

High-performance LLM inference engine forked from vLLM — adds advanced sampling (top-k, top-p, top-a, min-p, tau, mirostat), broader quantization (EXL2, GGUF, AQLM), Karras schedulers. Best for RP/creative writing pipelines. NVIDIA GPU 16GB+ VRAM.

LLM observability and tracing platform. OpenTelemetry-native. Capture, evaluate, and debug LLM application traces in real time.

Microsoft's visual builder for AutoGen multi-agent workflows — design agent teams, orchestrate conversations, run experiments. Web UI on top of AutoGen framework.

Stable Diffusion web UI — the most popular open-source image generation interface. Extensions ecosystem, ControlNet, LoRA training, inpainting, img2img. Requires NVIDIA GPU 6GB+ VRAM (8GB+ recommended for SDXL).

Conversational AI platform — visual flow builder, LLM-powered, multi-channel (web, WhatsApp, Slack, Telegram). Self-hosted alternative to Voiceflow + Rasa hybrid.

Embeddings database (vector DB) for LLM RAG and semantic search. Python-friendly, minimal footprint, easy integration with LangChain/LlamaIndex.

Robust blind face restoration via codebook lookup (NeurIPS 2022). More aggressive recovery than GFPGAN — handles severely degraded faces, AI-generated artifacts, color photos to B&W restoration. NVIDIA GPU 4GB+ VRAM.

Node-based Stable Diffusion + Flux workflow editor. Designed for advanced users running complex multi-stage image generation pipelines. Native support for Flux Dev/Schnell, SDXL, video models. NVIDIA GPU 8GB+ VRAM required.

Open-source AI code assistant for VS Code + JetBrains — autocomplete, chat, edit, agent mode. Connect any LLM (Ollama, OpenAI, Anthropic). The Cursor / GitHub Copilot alternative.

Visual builder for CrewAI multi-agent workflows — define roles, tasks, tools, then orchestrate as autonomous crews. Streamlit-based UI on top of CrewAI framework.

Open-source LLM app development platform — visual workflow builder, RAG pipelines, agents, prompt engineering, model management. The "platform" for AI applications.

IBM document conversion API. PDF, DOCX, PPTX, HTML → Markdown / JSON for RAG preprocessing. Layout-aware OCR, tables.

Low-code drag-and-drop LLM app builder. Visual flows for chatbots, agents, RAG pipelines. LangChain under the hood.

Midjourney-style easy-mode image generator built on Stable Diffusion. Smart prompt enhancement, hand-tuned defaults, minimal UI — first generation in 30 seconds with zero configuration. NVIDIA GPU 4GB+ VRAM (6GB+ recommended).

Practical face restoration by Tencent ARC — restores old/blurry/low-res face photos with GAN priors. Industry standard for face recovery. Works on faces from baby to elderly. NVIDIA GPU 4GB+ VRAM.

High-throughput embedding inference server — REST API for text + image embeddings, BGE/E5/Jina/Cohere model support, OpenAI-compatible /v1/embeddings endpoint. 5-20× faster than HuggingFace Inference. NVIDIA GPU 4GB+ VRAM.

Enterprise-grade Stable Diffusion UI — professional canvas editor, layer system, queue manager, multi-user model library. The mature alternative to AUTOMATIC1111 for production teams. NVIDIA GPU 8GB+ VRAM.

Personal AI assistant with RAG over your documents (Markdown, PDF, Notion, GitHub, Obsidian). Multi-LLM, web UI, mobile apps.

Open-source multilingual TTS model (~82M params). OpenAI-compatible API. Community Docker wrapper, multiple voices.

Visual LangChain pipeline editor. Build LLM agents, RAG flows, multi-step prompts via drag-and-drop. Export to Python or API.

Multi-provider chat frontend (OpenAI, Anthropic, Google, Ollama, Mistral, Bedrock). Multi-user, RBAC, file uploads, code interpreter.

Proxy/router for 100+ LLM providers (OpenAI, Anthropic, Bedrock, Ollama, Azure). Unified API, cost tracking, caching, rate limiting.

Large Language-and-Vision Assistant — open-source GPT-4V alternative. Multi-modal LLM that understands images + text. Variants: LLaVA-1.6/NeXT, LLaVA-OneVision (video). Industry leader in open vision-language models. NVIDIA GPU 16GB+ VRAM.

Premium ChatGPT UI with plugin marketplace, agent system, vision, voice, file analysis. Designer-quality interface, multi-LLM, MCP support. Built for the AI power user.

Drop-in OpenAI-compatible API for local inference. Runs LLMs, embeddings, audio, image gen on CPU. No GPU required.

Lightning-fast typo-tolerant search engine. Algolia alternative. Vector search support. Single-binary, low RAM, instant cold start.

Multi-agent framework that simulates a software company — agents play roles (PM, architect, engineer, QA) collaborating on tasks. From "one-line requirement" to working code.

Workflow automation with 400+ integrations + native AI nodes (OpenAI, Anthropic, Ollama). Zapier alternative for self-hosters.

ChatGPT-like web UI for any LLM provider — OpenAI, Anthropic, Google, Ollama, custom. Markdown rendering, code blocks, prompt library, multi-language. Most-starred LLM UI on GitHub.

No Language Left Behind by Meta AI — neural machine translation across 200 languages, including 150+ low-resource. Open weights, REST API for bulk translation. CC-BY-NC for non-commercial; CC-BY 4.0 for newer distillations. NVIDIA GPU 8GB+ VRAM.

Self-hosted LLM runtime (Llama 3, Mistral, Qwen, Phi). Local-first inference, CPU works, GPU optional. OpenAI-compatible API.

Lets LLMs run code locally — Python, JavaScript, Shell. Natural language → executable code → results. The "code interpreter for the rest of us", running offline.

ChatGPT-like web UI for local LLMs. Plugs into Ollama natively. Multi-user, RBAC, document RAG, web search.

Autonomous AI software developer (formerly OpenDevin). Reads, writes, debugs code, runs shell commands, browses the web.

Open-source search and analytics engine (Elasticsearch fork). Full-text + vector + observability search. Single-node mode.

High-performance vector database written in Rust. Powers RAG, semantic search, recommendation systems. REST + gRPC API.

Opinionated RAG Python library — ingest documents (PDF, DOCX, video, audio, code), build "second brains", plug any LLM (OpenAI, Anthropic, Mistral, Ollama). Quivr is now a Python SDK; this install ships a JupyterLab starter pre-configured with quivr-core.

Conversational AI framework. NLU + dialog management for chatbots and voice assistants. REST + WebSocket API.

Practical algorithm for general image/video upscaling — 2×/4× super-resolution, anime model variant, video frame-by-frame upscale. Used by Topaz alternatives. NVIDIA GPU 4GB+ VRAM (CPU fallback works for low-res).

Background removal API — strip backgrounds from any image (people, products, animals, hair). Multiple model variants (general, anime, human, dichotomous). REST API ready for high-volume batch. CPU works, NVIDIA GPU 4GB+ for batch speed.

Advanced fork of AUTOMATIC1111 by vladmandic — faster, broader model support (SD 1.5/2.x/3, SDXL, Flux, Stable Cascade), improved diffusers pipeline. Aggressive optimizations + active maintenance. NVIDIA GPU 6GB+ VRAM (8GB+ for SDXL/Flux).

Privacy-respecting metasearch engine aggregating 70+ engines. AI-friendly (used as web search tool for LLM agents). No tracking.

OpenAI-compatible STT + TTS server. Faster-Whisper + Kokoro/Piper bundled. Dynamic model loading per-request.

Autonomous AI agent framework — build, manage, run multi-agent workflows with tools, memory, and goal-driven loops. GUI + API. Self-hosted alternative to commercial agent platforms.

Modern document AI by Datalab — multilingual OCR (90+ languages), layout analysis, reading order detection, table recognition. Significantly higher accuracy than Tesseract on real-world documents. NVIDIA GPU 4GB+ VRAM (CPU OK for low volume).

Multi-backend Stable Diffusion orchestrator by Mcmonkey (Stability AI alumnus) — runs ComfyUI as backend, provides modern A1111-style UI, multi-GPU support, batch automation, model manager. Production-grade. NVIDIA GPU 8GB+ VRAM.

Self-hosted AI coding assistant (GitHub Copilot alternative). Code completion + chat. Open-source, runs locally on CPU.

Text Generation Inference by HuggingFace — production LLM serving with continuous batching, tensor parallelism, quantization (bitsandbytes/GPTQ/AWQ), streaming, OpenAI-compatible. The HF official serving stack. NVIDIA GPU 16GB+ VRAM.

Open-source chatbot builder with drag-and-drop flow editor. AI nodes for LLM integration. Embed in websites or share via link.

Lightning-fast typo-tolerant search engine. Vector + hybrid search built-in. Algolia alternative, native binary, no JVM.

High-throughput LLM inference engine — PagedAttention for memory efficiency, OpenAI-compatible API, tensor parallelism for multi-GPU. Best for production LLM serving. Serves Llama 3.x, Mistral, Qwen, DeepSeek, etc. NVIDIA GPU 16GB+ VRAM for 7B, 24GB+ for 13B.

Production-grade vector database (Go). REST + gRPC API. Powers RAG, semantic search, hybrid search. Alternative to Qdrant for larger-scale.

OpenAI Whisper speech-to-text webservice. CTranslate2 backend (faster-whisper) for low-latency CPU inference. REST + Swagger UI.

Production-grade ASR — Whisper inference 70× real-time with word-level timestamps, speaker diarization (pyannote), forced alignment. The standard for transcription + subtitling pipelines. NVIDIA GPU 6GB+ VRAM (4GB CPU mode OK).

State-of-art music source separation by Meta — split songs into vocals/drums/bass/other stems. Used by music producers, video creators, karaoke makers. Hybrid Transformer model. NVIDIA GPU 6GB+ VRAM (CPU OK for short audio).

State-of-the-art zero-shot TTS by Shanghai AI Lab — flow matching + DiT architecture, faster than XTTS-v2 with higher quality, 10-second voice cloning, English + Chinese natively. NVIDIA GPU 8GB+ VRAM.

Text-to-music generator by Meta AI (AudioCraft) — generate 30+ seconds of music from text prompts or melody conditioning. 3 model sizes (300M/1.5B/3.3B), genre/instrument/mood control. NVIDIA GPU 8GB+ VRAM (16GB for large).

Instant voice cloning by MyShell — clone any voice from a 6-second sample, generate speech in multiple languages with that voice + custom emotion/accent/rhythm. Faster + more flexible than XTTS. NVIDIA GPU 8GB+ VRAM.

Open-weight text-to-audio by Stability AI — generates 47-second clips of stereo audio at 44.1kHz from prompts. Specialized in sound effects, foley, short music samples. NVIDIA GPU 8GB+ VRAM.

Multilingual text-to-speech by Coqui AI — 17 languages, voice cloning from 6-second sample, expressive emotional delivery, streaming output. Industry-leading open TTS. NVIDIA GPU 6GB+ VRAM (4GB CPU fallback OK).

Real-time portrait animation by Kuaishou — drive a still image with another person's face video (expression + head motion transfer). Higher quality + faster than SadTalker. NVIDIA GPU 8GB+ VRAM.

Single image + audio → animated talking head video. Realistic 3D-aware face animation, head pose, expression. Generates 30 FPS portrait video from one photo + voice clip. NVIDIA GPU 8GB+ VRAM.

Self-hosted RSS feed aggregator. PHP-based, fast, supports thousands of feeds. Mobile-friendly, API for clients (Reeder, NetNewsWire, NewsBlur).

Self-hosted bookmark manager with AI-powered tagging, full-text search, link archival. Formerly Hoarder. Modern alternative to Pocket / Pinboard.

Minimalist self-hosted bookmark manager. Tags, full-text search, browser extensions, REST API. Single Django binary, SQLite by default.

Minimalist Go-based RSS reader. Single binary, Postgres backend, sub-100ms responses. Strong API support for mobile clients. Privacy-focused.

Self-hosted read-it-later — save articles, strip ads + sidebars, read distraction-free on web + mobile. Open-source Pocket / Instapaper alternative.

Federated link aggregator — Reddit-style communities over ActivityPub. Federate with other Lemmy instances + Mastodon + Pleroma. Rust backend.

Modern forum software on Node.js — real-time discussions, mobile-first, infinite scroll, plugin ecosystem. Self-hosted alternative to Discourse with simpler ops.

Open platform for developer portals — service catalog, tech docs, software templates, plugin ecosystem. Originally Spotify, now CNCF incubating.

Low-code platform for building internal tools, admin panels, dashboards from any DB or API.

Open-source backend-as-a-service — auth, database, storage, functions, real-time, messaging. Strong mobile + web SDK support. Firebase alternative.

Low-code platform for building business apps + automations. Drag-drop UI + integrations.

Open source status page system (Statuspage.io alternative). Incidents, components, scheduled maintenance.

Modern web server with automatic HTTPS via Let's Encrypt. Reverse proxy + static hosting.

Self-hosted Heroku/Vercel alternative — deploy apps, databases, services. Single-command install.

Headless CMS + data platform. Instant REST + GraphQL API on any SQL database. Admin UI, RBAC, webhooks, flows. Supabase-lite.

Container runtime engine. Installs Docker CE + docker-compose plugin via upstream get.docker.com.

Mini Heroku — git-push deploys, buildpacks, automatic SSL, scaling via Docker. Long-running OSS PaaS, alternative to Coolify with stronger Heroku compatibility.

Container-native CI/CD pipeline tool. Connects to Gitea/Gitlab/Github.

Encrypted backup to cloud (S3, Backblaze, Google Drive, etc.). Web UI + scheduler + dedup.

Lightweight MQTT 5.0 broker from Eclipse Foundation. Industry standard for IoT messaging. WebSocket support, TLS, persistence.

Enterprise-grade MQTT 5.0 broker. Handles 100M+ concurrent IoT connections. Web dashboard, rule engine, bridges to Kafka/databases.

Self-hosted Git service. Active community fork of Gitea, GitLab-like features. Active issues, PRs, wiki, releases, packages.

Lightweight Git service (GitLab alternative for smaller setups). Docker deploy with SQLite or PostgreSQL.

Self-hosted Git, CI/CD, container registry, issue tracking. Omnibus package install.

Observability dashboards. Visualize metrics, logs, traces from Prometheus, Loki, InfluxDB, Elasticsearch, and 60+ data sources.

Log aggregation system, Prometheus-inspired. Push-based ingestion via Promtail/Fluent Bit. Cost-effective alternative to ELK.

Instant GraphQL API on PostgreSQL. Authz/RLS, subscriptions, webhooks, REST endpoints. Production-grade for product teams.

Open-source API development ecosystem. Postman alternative — REST, GraphQL, WebSocket, SSE, MQTT, with collections and teams.

Industry-standard CI/CD automation server. 1900+ plugins, pipeline-as-code (Groovy DSL), distributed builds. Battle-tested since 2011.

Auto-discovering SNMP-based network monitoring (Observium fork). Alerts, billing, traffic graphs.

Standalone MariaDB 11 server with remote access enabled. MySQL-compatible.

S3-compatible object storage server. Single-node deployment via Docker.

WireGuard-based mesh VPN with identity, IAM, and policy. Self-hosted Tailscale alternative — connect devices/sites securely without managing peers manually.

Source-of-truth for network infrastructure. DCIM + IPAM. Track racks, devices, cables, VLANs, IP space, circuits. REST + GraphQL API.

Real-time per-second monitoring + alerting. Beautiful dashboards out-of-box.

Web UI for managing Nginx reverse proxy + Let's Encrypt SSL. Multi-domain routing on one VPS.

Visual flow-based programming for IoT, automation, APIs. Drag-drop nodes connect HTTP, MQTT, DB, files. 3000+ community nodes.

Open-source backend in 1 file. SQLite + auth + realtime + file storage + admin UI. Tiny footprint (~50MB), perfect for MVPs.

Lightweight web UI for managing Docker containers, images, volumes, networks.

Standalone PostgreSQL 16 server with remote access enabled.

Authoritative DNS server. RFC-compliant, DNSSEC, modern API. Used by major hosters + ISPs. Pair with PowerDNS-Admin for web UI.

Time-series database + monitoring system. Pull-based metric scraping. PromQL query language. Industry standard for cloud-native observability.

Game server management panel. Manages Minecraft, ARK, CS, Rust + 50 more game servers.

Self-hosted GIS server: QGIS Server backend (WMS/WFS/WCS/WMTS) + Lizmap web client frontend via NGINX + Redis cache + Postgres meta. The open-source standard for serving QGIS Desktop projects on the web.

In-memory data store / cache. Standalone Redis 7 with password-protected remote access.

Open-source Firebase alternative — Postgres database + auth + storage + realtime + edge functions, all in one platform. Major Firebase / Vercel-style BaaS competitor.

Low-code platform (Retool alternative). Drag-drop internal app builder, 50+ data sources.

Cloud-native reverse proxy + load balancer with automatic HTTPS via Let's Encrypt.

Self-hosted Uptime Robot — modern UI for HTTP/TCP/DNS/Docker container monitoring + status pages.

Lightweight private npm proxy registry.

Container-native CI/CD pipelines (community Drone fork). Simple YAML config, Gitea/GitHub/GitLab OAuth, multi-arch agents.

Developer-friendly API for PDF generation — HTML → PDF, Markdown → PDF, Office docs → PDF, image conversion. Powered by Chromium + LibreOffice headless.

Locally-hosted web-based PDF toolkit — merge, split, rotate, OCR, sign, compress, convert, watermark. 50+ tools in one Docker container.

Open-source invoicing platform — clients, estimates, invoices, payments, multi-currency. Laravel-based, customizable. Self-hosted alternative to InvoiceNinja for smaller setups.

Open-source IT asset management — track laptops, monitors, licenses, accessories, consumables. Checkin/checkout workflows, depreciation, audit reports.

Open-source Kubernetes web UI by Kinvolk/Microsoft — multi-cluster dashboard, customizable views, plugin system. Lens-alternative with no vendor lock-in.

Lightweight Kubernetes distribution — single binary, embedded etcd or external DB, runs Kubernetes workloads on edge / IoT / small clusters. CNCF Sandbox.

LLM fine-tuning toolkit by OpenAccess AI Collective — config-driven LoRA/QLoRA/full-parameter training, multi-GPU, DeepSpeed/FSDP, broad model support (Llama, Mistral, Qwen, Gemma). Industry standard for production fine-tuning. NVIDIA GPU 24GB+ VRAM.

Visual web UI for LLM training — supports SFT, DPO, ORPO, PPO, GRPO across 100+ models (Llama, Mistral, Qwen, ChatGLM, Phi, Gemma). No-code config builder, integrated dataset converter, training monitoring. NVIDIA GPU 16GB+ VRAM.

2× faster LLM fine-tuning with 50% less memory — custom Triton kernels, dynamic 4-bit quant, native multi-GPU. Maintained by Unsloth AI (Daniel Han, ex-Microsoft). Best for budget GPU fine-tuning. NVIDIA GPU 8GB+ VRAM (for 7B models in QLoRA).

Self-hosted newsletter and mailing list manager. PostgreSQL backed.

Dockerized mail server suite (Postfix, Dovecot, SOGo, ClamAV, Rspamd). Production-ready, single-domain or multi-domain.

Modern, full-featured mail delivery platform (transactional + bulk). Self-hosted SendGrid alternative.

Modern webmail (IMAP/SMTP client). Connect to external mail servers.

Modern, fast PHP webmail (RainLoop fork).

All-in-one Rust mail server (SMTP + JMAP + IMAP). Modern, secure, single-binary.

Self-hosted live video streaming + chat. RTMP-in / HLS-out. Twitch/YouTube Live alternative for creators who own their audience.

Companion to Sonarr/Radarr that downloads subtitles automatically. Supports OpenSubtitles, Addic7ed, custom providers. Multi-language.

Modern web UI for Calibre ebook libraries — browse, read in-browser, send to Kindle, OPDS feed for mobile readers. Replaces Calibre desktop content server.

Request manager for Jellyfin users — fork of Overseerr with Jellyfin authentication. Friends/family request, Sonarr/Radarr fulfill automatically.

Fast self-hosted ebook + comic reader/server — supports EPUB, PDF, CBZ/CBR, plain text. Cross-platform readers, library stats, reading progress sync.

Self-hosted comics & manga server — CBZ/CBR/PDF/EPUB libraries, OPDS, Kavita/Tachiyomi compatible. Browse and read in-browser with library management.

PVR for music — monitors artist releases, fetches albums, organizes library with MusicBrainz tagging. Music counterpart to Sonarr/Radarr.

Request management + media discovery for Plex users — friends/family request content, Overseerr forwards to Sonarr/Radarr automatically.

Federated video platform — host your own YouTube, federate with thousands of instances via ActivityPub. P2P streaming (WebTorrent) lowers bandwidth cost.

Centralized indexer manager for the *Arr stack — configure indexers once, push to all *Arr apps. Replaces Jackett with native protocol.

PVR for movies — monitors releases, grabs, organizes. Movie counterpart to Sonarr. Central piece of the *Arr media stack.

PVR for ebooks + audiobooks — monitors author releases, grabs new titles, organizes library. Books counterpart in the *Arr stack.

PVR for TV shows — monitors RSS feeds for new episodes, grabs releases, organizes the library. Hub of the *Arr media automation stack.

Open-source Notion + Miro alternative. Docs, whiteboards, databases in one unified workspace. Real-time multiplayer.

Open source accounting software for small businesses. Multi-currency, multi-company.

Enterprise BI platform from Apache. Rich dashboards, SQL Lab, 50+ data sources, role-based access. Tableau alternative.

Open-source Notion alternative — docs, databases, kanban, calendar, all-in-one workspace. Rust + Flutter native apps + self-hosted cloud sync via AppFlowy Cloud.

Self-hosted audiobook + podcast server. Mobile apps, ePub support.

Open-source no-code database (Airtable alternative). Spreadsheet-like UI, forms, kanban, gallery views, REST API.

Open-source web conferencing system built for online learning. HD video, screen sharing, breakout rooms, polls, whiteboard, recordings. Tight Moodle integration.

Self-hosted wiki/documentation platform. Docker deploy with MariaDB.

Open-source Intercom alternative — multi-channel customer support (live chat, email, SMS, WhatsApp).

Open-source document signing (DocuSign alternative). Send + sign PDFs with audit trail, multi-signer flows, branded templates.

Self-hosted Matrix homeserver (Synapse) + Element web client. Secure E2E-encrypted messaging.

Full-stack ERP (Accounting, CRM, HR, Manufacturing, Projects, Stock, Buying, Selling, Assets, Quality, Education, Healthcare, Agriculture).

Modern open-source CRM, lightweight + customizable. Sales pipeline, contacts, calendar.

Modern PHP forum platform. Lightweight, fast, mobile-friendly. Discussions, tags, mentions, real-time updates. Discourse-lite.

Self-hosted project management — kanban boards, table view, card properties. Built by Mattermost as an open-source Trello/Notion-board alternative.

Self-hosted Help Scout alternative — mail-driven shared inbox + ticketing.

Gamified habit tracker — turn your todos into an RPG with XP, gold, monsters, parties. Open-source self-hosted version of the popular productivity game.

Real-time collaborative Markdown editor — multi-user simultaneous editing of Markdown documents in browser, with revision history, public sharing, MathJax + Mermaid.

Self-hosted photo and video backup. Google Photos alternative. Mobile apps iOS/Android, auto-upload, ML face detection (optional).

Open-source invoicing + payments + recurring billing. Stripe/PayPal/Square integration.

Free media server (Plex/Emby alternative). Streams movies, TV, music, photos.

Self-hosted video conferencing platform. Docker-based deploy.

Self-hosted sync server for Joplin notes app. End-to-end encrypted notes across desktop + mobile + web. Markdown-native.

Lightweight project management with kanban boards. Minimalist, PHP-based.

Open-source knowledge graph — networked thought, daily journaling, bidirectional links, block references. Local-first markdown + org-mode. Roam Research / Obsidian alternative.

Self-hosted Slack alternative. Docker deploy with PostgreSQL.

Recipe management + meal planning. Scrape recipes from any URL, weekly meal plan, shopping lists, nutrition, multi-user households.

Lightweight self-hosted memo / microblogging hub. Markdown notes with tags, daily review, Telegram + RSS integrations. Privacy-first.

BI dashboards + SQL query tool. 25+ database connectors (PG, MySQL, MongoDB, ClickHouse, BigQuery). Self-service analytics.

Open-source learning management system (LMS) used by universities, schools, and corporate training programs worldwide. Courses, quizzes, grades, forums, video, mobile app.

Modern music server (Subsonic-compatible). Streams MP3/FLAC/Ogg, Last.fm scrobble.

Self-hosted file sync, collab, calendar, contacts. Docker-based deploy with built-in PostgreSQL + Redis.

Open-source ERP + CRM + accounting + inventory + e-commerce + manufacturing. 40+ business modules in one platform.

Self-hosted document editor (Word/Excel/PowerPoint compatible). Integration backend for Nextcloud/ownCloud/SharePoint.

Open-source project management (Jira alternative). Gantt, agile boards, time tracking.

Open-source customer support ticketing platform. PHP-based, the de-facto standard for hosting resellers.

Open-source team knowledge base / wiki (Notion alternative). Requires external SSO provider (Slack/Google/OIDC).

Document management with OCR, full-text search, tags, correspondents. Scan paperwork → searchable digital archive. Tika + Gotenberg for office docs.

Open-source design and prototyping platform (Figma alternative).

AI-powered self-hosted photo gallery (Google Photos alternative). Tagging, search, faces.

Open-source project management — issues, cycles, modules, pages. Jira alternative built for engineering teams. Self-hosted with Postgres + Redis + MinIO + nginx.

Open-source resume builder — drag-and-drop sections, multi-language, dozens of templates, PDF export. Self-hosted privacy-first alternative to Resume.io.

Query builder + visualization + dashboards for any SQL/NoSQL data source. Open-source alternative to Mode/Looker for BI teams.

Self-hosted team chat (Slack alternative). MongoDB backend.

Agile project management for Scrum/Kanban teams. Sprint backlog, taskboards, time tracking.

Self-hosted recipe manager — import from any cooking website, plan meals, generate shopping lists, multi-user households. Django + PostgreSQL stack.

Hierarchical note-taking with strong tree organization, scripting, encryption, sync. Power-user note app with depth Joplin/Logseq don't reach.

Modern Notion-inspired open-source CRM. Rich UI, customizable objects.

Open-source to-do app (Todoist alternative). Boards, kanban, tasks.

Open source CRM. Sales, marketing, customer service, project management.

Open-source kanban board (Trello alternative). MongoDB backend.

Modern wiki on Node.js — Markdown editor, Git-based sync, OAuth, 50+ auth strategies. Built for teams that outgrew Notion-like prison and want self-hosted control.

Industry-standard secrets management platform — store API keys, certificates, database credentials. Dynamic secrets, encryption-as-a-service, audit logging.

Modern open-source secrets manager — alternative to HashiCorp Vault with focus on developer UX. Secret rotation, dynamic secrets, audit logs, multi-env.

Team password manager — open-source alternative to 1Password Teams / Bitwarden Business. Strong on team collaboration, fine-grained sharing, audit log.

Network-wide ad blocker + DNS server. More modern alternative to Pi-hole.

Authentication and authorization server with 2FA / WebAuthn for reverse proxies.

Identity provider / SSO (Keycloak alternative). OIDC, SAML, LDAP. Docker deploy.

Modern Web Application Firewall (WAF) + reverse proxy. Nginx-based, ModSecurity rules, ClamAV scan, Let's Encrypt auto-SSL, bot detection.

Collaborative IPS/IDS — detects threats via log analysis + shares IPs with global blocklist.

End-to-end encrypted collaborative office suite. Docs, sheets, slides, kanban, forms — zero-knowledge server cannot read content.

Lightweight ActivityPub federated social server (Mastodon alternative). Single Go binary, SQLite default, no Redis/Sidekiq overhead.

Self-hosted, open-source Tailscale control server (DERP relay + coordination).

Enterprise-grade IAM / SSO platform. OIDC, SAML, social login, multi-tenant.

Open-source LDAP directory server with phpLDAPadmin web UI.

Network-wide ad blocker, DNS sinkhole. Docker-based.

Rust-based, lightweight Bitwarden-compatible password manager server. Docker deploy.

Open-source SIEM + XDR security platform. Log analysis, intrusion detection, compliance.

WireGuard VPN with web UI (wg-easy). Manage clients with QR codes.

Build custom firmware for ESP8266/ESP32 boards from YAML config. Web dashboard to compile, upload OTA, and monitor your IoT fleet.

Open-source NVR with real-time AI object detection. Powers smart home camera setups, integrates with Home Assistant. CPU works; Coral USB or GPU accelerates substantially.

World's leading open-source smart home platform. 2000+ integrations, automations, voice control, energy management, all running locally.

Web interface for controlling 3D printers — monitor prints, slice G-code, control via webcam stream. Standard 3D printing remote management.

Bridge Zigbee devices (sensors, switches, bulbs) to MQTT. Supports 3000+ Zigbee products. Decouples Zigbee fleet from any specific home automation platform.

Time-series database for IoT, monitoring, real-time analytics. High write throughput, downsampling, retention policies, Flux query language.

High-performance Prometheus-compatible TSDB. Lower memory + faster than Prometheus at scale. PromQL + MetricsQL. Drop-in replacement.

Personalized motion module that animates any SDXL/SD 1.5 checkpoint — generates 2-8 second videos using your existing checkpoints + LoRAs. Best for stylized animation. NVIDIA GPU 8GB+ VRAM.

Open-source text-to-video by THUDM/Zhipu AI — generates 6 seconds at 720×480 with strong temporal consistency. Multiple model sizes (2B, 5B). One of the highest-quality open video models. NVIDIA GPU 16GB+ VRAM (24GB+ for 5B).

Largest open-source video foundation model by Tencent (13B params) — cinema-quality text-to-video, image-to-video, 720p output, strong physics + camera motion. December 2024 release. NVIDIA GPU 24GB+ VRAM, 40GB+ recommended.

Real-time text/image-to-video by Lightricks — generates 5 seconds of video in ~4 seconds on a single GPU. State-of-the-art quality at fastest speed. Open weights, MIT license. NVIDIA GPU 12GB+ VRAM, 24GB+ recommended.

Modern Node.js blogging / newsletter platform. Docker deploy with MySQL backing store.

Self-hosted Google Analytics alternative. MariaDB backend.

Open-source Airtable alternative. Database front-end with REST API auto-generation.

Privacy-friendly Google Analytics alternative. Docker deploy with PostgreSQL + ClickHouse.

Open-source e-commerce platform. Lighter alternative to Magento, French-origin.

Modern headless e-commerce platform. GraphQL API + Next.js storefront. Python/Django.

Open-source headless CMS. REST + GraphQL APIs, admin panel.

Lightweight, privacy-friendly web analytics. Cookieless. PostgreSQL backend.

Open-source content management system powering 40%+ of the web. Installed with Nginx + PHP-FPM + MariaDB.