Локальные LLM
Jan
Jan is an open source alternative to ChatGPT that runs 100% offline on your computer
★ 44,8 тыс. звёздCustom (check repo)
Смотреть руководство →CoreCoder — проект с открытым исходным кодом из категории «ИИ-агенты» от he-yufeng (he-yufeng/CoreCoder на GitHub). У него 1 794 звезды на GitHub и 440 форков, он распространяется по лицензии MIT и последний раз обновлялся на GitHub 2 октября 2026.
Self-hosted ai agents: minimal ai coding agent (~1,000 lines of python) inspir.
Fetch the source code onto your machine.
Set up the Python environment the project needs.
Copy the sample config or env file and adjust it.
Start the app from the source directory.
Built around ai agent for a focused workflow.
Built around claude code for a focused workflow.
Built around cli for a focused workflow.
Built around coding agent for a focused workflow.
Inspect, fork, and modify every line of the code.
Run it on your own machine — no account lock-in.
Get the source on your machine.
git clone https://github.com/he-yufeng/CoreCoder.gitSet up Python dependencies.
cd CoreCoder && pip install -r requirements.txtLaunch the application.
cd CoreCoder && python main.pyIt minimal AI coding agent (~1,000 lines of Python) inspired by Claude Code. Works with any LLM. Think NanoGPT for coding agents. Formerly NanoCoder, giving you a working starting point you can run and adapt.
Yes — the code is open-source, so you can use and modify it without a license fee.
Clone the GitHub repository and follow the README; the typical flow is install dependencies, configure, and run.
It is primarily written in Python, so you will need that toolchain to build or extend it.
The GitHub repository is the hub: open issues for bugs and check the README and discussions for guidance.
Jan is an open source alternative to ChatGPT that runs 100% offline on your computer
High-performance Nano Banana image platform — Gemini power, up to 4K
Personal research assistant: lit reviews, notes, experiments, paper writing.
Rebuild Claude Code’s core in ~5,000 lines: an 11-chapter coding-agent masterclass.
Autonomous AI agents for penetration testing — self-hosted, multi-agent, with a full web UI
Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat…
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если CoreCoder вам не подходит.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Соседние проекты из категории «ИИ-агенты» в каталоге, ближайшие по числу звёзд к CoreCoder — продолжайте изучать полку по обе стороны от него.
Android GUI agent framework: AI that sees, plans, and taps real mobile apps.
Dev agent multiplexer designed for git worktrees and coding agents.
Management console for installing and monitoring AI agents and pipelines.
Reliable computer use for any agent: drive real apps through accessibility trees, not pixel guessing.
Babysitter enforces the process: deterministic, quality-gated orchestration for fleets of coding agents.
Retired agent framework: its prompts and guides live on in successor repos.
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.