Work with Sweet directly in your codebase. Plan, build, and ship from your terminal or your browser — same agent, either way.
Full terminal integration. Works alongside any IDE, any stack.
The same agent, no terminal required — reads and edits your real files in an isolated container.
Browser or terminal — same agent, same workspace, same history either way.
Plan → execute → verify → recover, over and over, until the work is done.
Runs your tests and reads the failures before it calls it done.
Flaky test? Wrong assumption? It diagnoses, retries, and keeps going.
Splits work across subagents that coordinate with each other.
Node, Python, Go, Rust, Rails — plus ML, scripts, and more.
Set a horizon and walk away — it keeps working until the timer expires.
Runs for hours, with subagents working in parallel.
Open-source models — more usage for your dollar.
Near-frontier intelligence at the best price-to-performance ratio available.
Every session now runs on 4 vCPUs and 8GB of RAM — enough headroom for a full build, not just a script.
The files menu now zips and downloads an entire folder in one click.
Tool calls collapse to a single line by default, and the reasoning trace lives behind its own dropdown — expand only what you want to inspect.
Give it a time budget and a goal. Sweet loops on its own — plan, execute, verify, recover — until the timer runs out.
Ship real tasks today. 25M tokens daily, resets every 24 hours.
For developers who ship with Sweet daily. 10× the free tier, unlimited loops.
For teams and heavy users who run Sweet around the clock.
npm install -g @sweet-cli/sweet, then sweet login.Free to start — open it in your browser, or drop into your terminal.
Open Web App →