Jenny is a free, open-source Electron desktop application for running local LLMs with tool calling, rollback, and IDE features, developed as a solo project over 1.5 years.
Hi all! I just wanna say that I am tired lol. Yes, it's another harness, but I spent a lot of time and effort and have forsaken my hobbies to build the Jenny (like XJ-9) app. Jenny is a free, MIT licensed electron desktop app for running local LLMs with tool calling, rollback, and an IDE. A lot of you probably had the same thought I did a year or year and a half ago: frontier LLM use is subsidized heavily by private equity and venture capital, which will eventually dry up and then be enshitified. So, I started building a harness that can host local LLMs privately. I went through many vibecoded iterations throughout the past 1.5 years and finally landed on a native electron desktop app. I wanted to make it easy and seamless for people. I am not a professional software dev but I do have a deep personal interest for it and AI tech. HOWEVER, the Jenny project became a second job and its in a state that I think it's ready for release. This has been a solo project and super fun. I know there are many other options out there that beat me to the punch like Unsloth Desktop (wonderful btw), LM Studio, and Open WebUI, but I hope someone can enjoy Jenny and what it has to offer! I plan to maintain and improve the app, but again, solo here and I do have a day job and friends that I should focus on a bit more. (Opening issues are welcomed, but please be gentle!) Some highlights: - Private and locally run, no network calls except to your local model runtime (only private user facing telemetry so you can debug and troubleshoot) - Fully open source, MIT License - Fun and pretty chat UI (imo) - Makes small models capable (highly recommend ornith1.5:9b for tool calling and speed!), but with safety and rollback features so WHEN a small model screws up bad, you don't have to worry. Destructive shell commands need approval and file edits are checkpointed! - Full IDE, for you handcrafted code enjoyers - Some assistant like features like calendar and scratchpad that the model is able to read/modify - Data rich diagnostics and logs - llama.cpp, vLLM, or any OpenAI-compatible local endpoint, plus GGUF via the managed llama-server (for MTP) I would really appreciate any feedback to validate the time and mental pain that was put into this project. I whine, but it's all love! I hope Jenny helps you build too. Windows build is solid (performance too with 5070ti and ornith15:9b), MacOS is untested cause I'm a poor. No Linux (sorry, but blame engine providers) https://github.com/SaltyPretz3l/jenny I also made an unsigned installer .exe for convenience, but understand if you don't trust it! SmartScreen warning will appear. https://github.com/SaltyPretz3l/jenny/releases/latest/download/Jenny-Setup-x64.exe
The author introduces Vellium, an open-source cross-platform desktop application for interacting with LLMs, featuring new desktop widgets and a visual interface for AI agents that support MCP servers and file manipulation.
A Windows desktop app called SEELS that allows users to run local LLMs, correct model responses, and automatically train LoRA adapters from corrections. It bundles voice mode, hardware dashboard, and a training sidecar.
Juggler is an open-source GUI coding agent that provides a visual workbench for interacting with LLMs to edit code, featuring inspectable tool calls, branching threads, and editable context. It runs locally or remotely and supports multiple models.
Open WebUI Desktop launches as a native app letting users run local LLMs or connect to remote servers without Docker or terminal setup, featuring offline operation, system-wide voice input, and floating chat overlay.