Open source AI tools
A curated set of open source AI tools that run on your own server: model front-ends, recognition, video and audio processing, agents. Every entry is described in plain language.
Open source AI here means finished tools you deploy yourself: chat front-ends for local models, speech and text recognition, image, video and audio processing, search and agent systems. Your data stays on your own server, and usage of the model is not billed by a third-party service.
Only complete products get in, and only ones with a way to run them: Docker, a ready interface, a command line or an API. Developer libraries, link lists and bare model weights are left out, because you cannot launch them and get a result. Each card explains what the tool does, who it suits and what you will need to run it. The repository link is unlocked from the Creator plan (also on Make, N8N and Full).
GeoLook — GEO Optimization Platform for AI Search
GeoLook is an open-source self-hosted platform for Generative Engine Optimization (GEO) that implements a full-cycle pipeline for improving a site's visibility in AI search engines: status analysis, d
Cloudroom Core — Cloud Runtime for AI Coding Agents
Cloudroom Core is a self-hosted Rust runtime that lets you run Claude Code, Codex, and other AI coding agents in the cloud on your own server without relying on paid managed services. You need it when
AppFlowy — рабочее пространство с ИИ и локальными данными
AppFlowy is an open-source alternative to Notion: an AI-powered workspace for notes, documents, databases, tasks, and projects where you don't lose control of your data. Reach for it when you want a k
frp — проброс локальных сервисов через NAT в интернет
frp (fast reverse proxy) is a self-hosted tunnel that exposes a service trapped inside a private network (behind NAT, a home router, or a firewall) to the public internet when you have no static IP. R
Aurelio — Self-Hosted Personal Finance with AI Advisor
Aurelio is a self-hosted personal finance app with a built-in LLM-powered AI financial advisor that lets you track your net worth, investments, ETFs, cash, and debts entirely on your own computer. Use
DeepSeek Harness — плагинный каркас для ИИ-агентов
DeepSeek Harness (dsh) is an open-source agent harness from DeepSeek AI built on an "everything-is-a-plugin" architecture powered by the Cordis framework. Reach for it when you want an extensible envi
GitHub Readme Stats — карточки статистики для профиля
github-readme-stats is an open-source tool that renders GitHub profile statistics as cards you embed in a README. Reach for it when a profile or project page needs presenting: a card with commits, sta
Strapi — headless CMS на JavaScript
Strapi is an open-source headless content management system: it stores content and serves it over an API while your application decides how it looks. Reach for it when the same content must appear in
Local AI video dubbing with voice cloning
dub-studio is an open-source offline AI video dubbing studio for Windows that re-voices any video into another language — cloning the original speaker's timbre, translating captions, and localizing on
Tabby — современный терминал для Windows, macOS и Linux
Tabby is an open-source terminal for Windows, macOS and Linux with built-in SSH, Telnet and serial clients. Reach for it when the stock console gets in the way: you want tabs and split panes, saved se
Gemini CLI — ИИ-агент в командной строке
Gemini CLI is Google's open-source terminal AI agent. Reach for it when the command line is the comfortable place to work: get oriented in an unfamiliar project, edit code, write a script, process fil
ClickHouse — колоночная база для аналитики
ClickHouse is an open-source column-oriented database for analytics that computes reports over enormous volumes fast. Reach for it when an ordinary database stops coping with questions like "how many
Hermes Agent Russian Localization
Concat — offline video editor without watermarks
Concat is an open-source desktop video editor built as a free, self-contained replacement for CapCut — no watermarks, no account, no cloud uploads. It is written in Rust and ships as a native applicat
MiniMax Music — local AI music generation
MiniMax-Music3-Studio is an open-source desktop studio for generating full-length AI music tracks locally on your own GPU — no Python, no Node.js, no launcher, just a single Windows executable. Reach
2.5D AI Agent Office on Your Desktop
bagidea-office is an open-source tool that turns your desktop wallpaper into a living 2.5D AI-agent office built on Claude Code — agents walk to their desks, hold meetings, learn new skills, and accum
AutoGPT — платформа автономных ИИ-агентов
AutoGPT is an open-source platform for building autonomous AI agents that carry out multi-step tasks unattended. Reach for it when a recurring job should be handed over whole: monitor a topic and comp
Maestro — local AI studio for video and music clip creation
Maestro is an open-source local AI creative studio that handles the full video production cycle — music videos, short films, image and audio generation, and timeline editing — entirely on your own har
VoiceStudio — local voice cloning and speech synthesis
VoiceStudio is an open-source tool for voice cloning, text-to-speech synthesis, dictation, video dubbing, and audiobook production that runs entirely on local hardware — no cloud, no account, no subsc
Voicebox — локальная студия синтеза и клонирования голоса
Voicebox is an open-source voice studio that runs locally: voice cloning, text-to-speech and dictation in one application. Reach for it when you want to narrate a video, a course or a podcast in your
OmniRoute — Unified AI Gateway for 1200+ Models
OmniRoute is a self-hosted, MIT-licensed AI gateway that exposes a single API endpoint across 352+ providers and 1200+ language models, including Claude, GPT, Gemini, DeepSeek, Kimi, Qwen, GLM, and Mi
Auto-clipping long videos into viral Shorts and Reels
autoshorts is an open-source desktop application that automatically cuts long video or audio recordings into high-impact vertical short-form clips (9:16 portrait) with AI-powered viral moment ranking.
CrisperWhisper — accurate speech transcription with timestamps
CrisperWhisper is an open-source speech-to-text library that offers controllable verbatim or clean-intended transcription, precise word-level timestamps, and multilingual support across most languages
Experiential — Self-Hosted BYOK LLM Gateway
Experiential is an open-source BYOK LLM gateway with zero markup that learns from real traffic: automatically reduces costs, recommends better-performing models, and helps you fine-tune a specialized
FAQ
- What do I need to run one of these?
- Most often a server with Docker. Tools that work with video or large models need a GPU, and the description says so.
- Does open source mean free?
- The code is free. You still pay for the server, and sometimes for outside model access if the tool calls a cloud model.
- Why self-host instead of using a cloud service?
- Data never leaves your infrastructure, cost does not scale with request volume, and nobody can switch the tool off from outside. In exchange you take on updates and uptime.
- Why are there no libraries or awesome lists?
- You cannot get a working result out of them without writing software. This catalog collects things that run and pay off right after deployment.