ローカルLLM
helmor
A local workbench for multi-agent software development.
★ 1.3k スターApache-2.0
ガイドを見る →agent-sessionsは、jazzyalexが公開しているオープンソースのローカルLLMプロジェクトです（GitHub: jazzyalex/agent-sessions）。GitHubスターは889、フォークは65で、MITライセンスの下で公開されています。GitHubでの最終更新は2026年10月2日です。
Browse and resume your AI coding-agent sessions on macOS.
Decide what you want the project to do before installing anything.
Grab the source and skim the README for setup requirements.
Install dependencies and add any API keys or settings it needs.
Launch it and follow the example workflow from the docs.
Orchestrates several local agents at once.
Works with Ollama, llama.cpp, and more.
Browse and resume past agent sessions.
Your code and prompts never leave your machine.
Runs on macOS, Linux, and Windows.
Designed to be practical on consumer hardware.
Download the source and move into the project folder.
git clone https://github.com/jazzyalex/agent-sessions.git
cd agent-sessionsCompile with the Swift toolchain.
swift buildLaunch the built binary.
.build/debug/AppIt is an open-source project at jazzyalex/agent-sessions with 65 forks. Check the repo’s contributors and recent commits to judge how active maintenance is.
Yes. It is MIT licensed, so you can use it commercially with attribution.
A modern laptop works for smaller models; bigger models want more RAM or a GPU.
Usually anything Ollama or llama.cpp can run — check the README for tested models.
Yes — local inference means prompts never leave your machine.
A local workbench for multi-agent software development.
ChatGPT for your home, running on a Raspberry Pi.
The open source coding agent
Write an agent skill once, sync it to 32 AI coding tools.
Agent! now supports macOS 14.6 or later, Apple Silicon and Intel, latest release.
Local-first dashboard of tokens, models, and cost for coding agents.
LibreAIカタログに掲載されたローカルLLMプロジェクトの中で最もスターの多いもの — agent-sessionsが合わなかった場合に検討したい、現実的な候補リストです。
The simplest way to run open language models on your own machine.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
LLM inference in C/C++
A high-throughput and memory-efficient inference and serving engine for LLMs
🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
カタログの中でagent-sessionsとスター数が近いローカルLLMプロジェクト — 前後の棚もそのまま見て回れます。
A talking LLM voice assistant that runs fully offline on your computer.
Early proposal and implementation for connecting LLMs to web pages.
One-command installer for a full self-hosted AI stack on your server.
Local security auditor for AI API relays — catches injection and swaps.
Run siliconscope locally: sudoless apple silicon system monitor (native swiftui g.
Desktop AI agent platform with one-click models and offline-first RAG.