Home/Void
Void open-source AI code editor logo

Void

Void is an open-source, VS Code-based AI code editor that offers autocomplete, inline edits, chat and agent modes, direct connections to cloud model providers, and support for locally hosted private models without routing prompts through a proprietary backend.

Visit website
Void VS Code-based private AI development environment

Product overview

What is Void?

Void is an open-source AI integrated development environment built on VS Code. It combines familiar editor and extension workflows with AI autocomplete, inline Quick Edit, normal chat, Gather Mode, and Agent Mode. Developers can connect directly to commercial model providers or run supported open models through local and self-hosted systems, keeping control over which provider receives code and prompts.

The editor advertises checkpoints for AI changes, lint-error detection, fill-in-the-middle model support, fast application of edits, and work on large files. Direct provider connections reduce one intermediary but do not make every workflow private: cloud models still receive submitted context, and local models, extensions, terminals, repositories, and tool calls retain their own risks. Review diffs and permissions before accepting agent actions.

How to Use Void

  1. Download the beta for a supported operating system and open a trusted repository.
  2. Choose a local, self-hosted, or cloud model and configure credentials securely.
  3. Use Tab for reviewed completions and Quick Edit for bounded inline changes.
  4. Use chat or Gather Mode to inspect code before assigning an Agent Mode task.
  5. Review checkpoints and diffs, then run formatting, tests, type checks, and security checks.
  6. Commit only after confirming the change scope and removing exposed secrets.

Core Features

  • AI autocomplete: Suggests code that can be applied with Tab.
  • Inline editing: Rewrites selected code from natural-language instructions.
  • Chat and agent modes: Supports questions, context gathering, and multi-step code changes.
  • Bring any model: Connects to major hosted providers and OpenAI-compatible endpoints.
  • Private model support: Works with local or self-hosted model runtimes such as Ollama and vLLM.
  • Change checkpoints: Helps developers inspect and recover from AI modifications.

Use Cases

  • Complete repetitive code while retaining editor-level review.
  • Refactor a bounded function or fix lint errors.
  • Explore an unfamiliar repository through code-aware chat.
  • Use a local model for repositories with stricter data controls.

Pricing

Void is open source and its beta can be downloaded without a software subscription. Users pay any infrastructure or API charges for the models they connect, including cloud-provider tokens or local hardware and hosting. Availability and maturity should be assessed as beta software.

Frequently Asked Questions

Does Void send prompts through its own model backend?

The site says it connects directly to providers rather than routing messages through a proprietary intermediary. The selected cloud provider may still process the submitted code.

Is local inference automatically secure?

No. Users must secure the machine, model server, extensions, repositories, logs, credentials, and backups.

Can agent changes be merged without review?

They should not be. AI edits can introduce bugs, insecure behavior, dependency risks, and incorrect assumptions.

Back to product directory

Related products

OpenAI Codex is an AI coding product for delegating scoped repository tasks and reviewing the resulting changes.

Claude Code is Anthropic’s AI coding product for iterative repository work through supported development surfaces.

Gemini CLI is Google’s open-source AI agent for terminal workflows and local development tasks.

Newsletter

Keep up with useful AI products

Get a concise selection of new products, practical use cases, and builder updates.