Void - open source Cursor AI code editor
Hey HN, we're Andrew and Mat and we're building Void, an open source Cursor alternative. Our editor is an open-source VSCode fork. Our primary goal is to create a fully-featured AI editor where you can keep your data completely private. Going open source is really important because it let us take the logic that would normally live on a server, like prompt building or message handling, and run it on your local machine instead. This way you can do things like self-host Llama 3.1 or Gemma and never have your data leave your computer. You can also just send your prompts directly to foundational…
In plain words
Void is an open-source code editor built on VSCode that integrates AI capabilities while prioritizing data privacy. Users can run AI features entirely locally using models like Llama 3.1 or Gemma, or connect directly to services like OpenAI and Anthropic without intermediary APIs. The open-source approach allows logic typically confined to servers to run on users' machines, enabling full control over prompts, message handling, and filesystem indexing. It is designed for developers who want AI-assisted coding without compromising data privacy.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN, we're Andrew and Mat and we're building Void, an open source Cursor alternative. Our editor is an open-source VSCode fork. Our primary goal is to create a fully-featured AI editor where you can keep your data completely private. Going open source is really important because it let us take the logic that would normally live on a server, like prompt building or message handling, and run it on your local machine instead. This way you can do things like self-host Llama 3.1 or Gemma and never have your data leave your computer. You can also just send your prompts directly to foundational models (GPT4o, Claude) without going through a middleman API. Another exciting part of the project is that we have full control over the IDE, and can build our own native features. Normally if you want to build an AI tool you have to make the same chat UI, history management, filesystem indexing, LSP, VSCode integrations, etc. We think we can standardize all these into a nice native API you can just plug into instead. It would be awesome to create a centralized place that lets people build advanced open source tools like AI edits across many files, and even agents, in an IDE we're all used to. Let us know what you think of the idea. We've really enjoyed working on it and are super excited to see where it goes :)
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, September 2024
the whole month →
BeforeSunset AI 2.0▲1,267Personalized AI daily planning that suits your life
AI · 2024 · beforesunset.ai
