Void

Void is an open-source AI code editor forked from VS Code that prioritizes data privacy, allowing developers to connect their own LLM providers or run models entirely locally so that code never leaves their machine. It positions itself as the privacy-conscious alternative to Cursor and other commercial AI editors that route code through third-party servers. Void provides the core AI coding features that developers expect from a modern AI editor: intelligent code completions, an AI chat panel with codebase context, inline code editing and generation, and the ability to ask questions about your code. Because it is built on the VS Code codebase, Void maintains full compatibility with VS Code extensions, themes, and keybindings, which means developers can switch to Void without losing their existing setup. The key difference is in how AI features are powered. Instead of requiring a subscription to a specific service, Void lets you bring your own API key from providers like Anthropic, OpenAI, or others, or connect to locally running models through Ollama or similar local inference servers. The privacy advantage is Void's primary differentiator. For developers working on proprietary codebases, in regulated industries, or at organizations with strict data governance policies, the ability to keep all code local while still benefiting from AI assistance addresses a real concern. When using local models, no code or context leaves the developer's machine at all. When using cloud APIs, the developer has a direct relationship with the API provider and full visibility into what data is being sent, rather than relying on a third party's data handling promises. Void is best suited for privacy-conscious developers, teams at security-sensitive organizations, and open-source enthusiasts who prefer transparent, community-driven tools. It works well for developers who already have API keys or the hardware to run local models. The editor is completely free to use, with costs limited to whatever API usage you incur from your chosen provider. As an early-stage open-source project, Void is less polished than Cursor in terms of AI experience quality, and running capable local models requires substantial hardware -- typically a machine with a modern GPU and at least 16GB of VRAM for a model that provides useful coding assistance.