Gray vs Continue: which should you use?

Last updated

Short answer

Choose Continue if you want AI inside your editor — autocomplete, chat and agent mode in VS Code or JetBrains. Choose Gray if you want a minimal MIT terminal harness that runs agent work and scheduled jobs outside any editor.

GrayContinue
What it isMinimal, always-on AI harness: one Rust binary, open source (MIT)Open-source (Apache-2.0) AI extension for VS Code and JetBrains
Where it worksThe terminal, and as a background service with the gatewayInside the editor: autocomplete, chat, edit and agent modes
ModelsAny OpenAI-compatible provider with your own API key; local models through Ollama; Claude Pro/Max and ChatGPT plans through official pluginsModel-agnostic by config — local and hosted providers, plus shared configs through its hub
Extends withPlugins and skills; MCP servers through the official pluginConfig-defined blocks: models, context providers, MCP tools, rules
Runs 24/7Yes — cron and app connections with no terminal openRuns while your editor is open
PriceFree (MIT); you pay your model provider. Hosted Voyager plan $77/month, not yet openFree (Apache-2.0); optional hub services

Is Gray an alternative to Continue?

Different surfaces. Continue puts AI inside the editor you're already in; Gray is a terminal agent and always-on harness. If you want completions and chat in the IDE, Continue is the fit; if you want an agent that keeps working when the editor is closed, Gray.

Do both support local and custom models?

Yes. Continue is configured per-model in its config; Gray takes any OpenAI-compatible provider or Ollama out of the box; see Providers & models.

When should you choose Continue instead?

Choose Continue for in-editor autocomplete plus agent work and shared team configs through its hub.

Sources

Competitor details are taken from the sources above as of the date at the top of this page. Products change quickly — if something here is out of date, open an issue on GitHub.