What it does
Supermaven suggests code as you type and uses definitions found elsewhere in your codebase to shape its suggestions. It is available for VS Code, JetBrains IDEs and Neovim. The site's table lists a tool latency of 250 ms against 783 ms for an unnamed leading competitor, and says it was the first completion tool to offer a 1 million token context window, which is a Pro and Team feature. Supermaven Chat is a chat interface for developers. It lets you pick between models listed on the site, including GPT-4o, Claude 3.5 Sonnet and GPT-4, and uses hotkeys to view diffs, apply changes, start conversations and switch models. You can attach recently edited files, and a single click sends your code together with compiler diagnostics so a model can help fix errors. There are three plans. The free plan covers fast suggestions and large codebases, with a 7-day data retention limit. Pro adds style adaptation, the 1 million token window, the site's largest model and $5 a month in chat credits, and includes a 30-day free trial. Team costs the same per user and adds centralised user management and billing.
Capabilities
- Inline code suggestions for VS Code, JetBrains IDEs and Neovim
- Contextual suggestions that draw on definitions elsewhere in the codebase
- 1 million token context window on Pro and Team
- Chat with GPT-4o, Claude 3.5 Sonnet, GPT-4 and other models, with diff and apply hotkeys
- Free plan, with a 30-day free trial of Pro
When to use it
- You want fast completions inside VS Code, a JetBrains IDE or Neovim
- You work in a large codebase and want a long context window
- You want to start on the free plan or the Pro trial before paying
