Windsurf drops SWE-1 — their first family of software engineering models

Similar to @Claude by Anthropic Code and @Codex by OpenAI — @Windsurf has announced SWE-1, its first family of software engineering models:

• SWE-1: Approximately Claude 3.5 Sonnet levels of tool-call reasoning while being cheaper to serve. It will be available to all paid users for a promotional period of 0 credits per user prompt.

• SWE-1-lite: A smaller model that replaces Cascade Base at better quality. It is available for unlimited use to all users, free or paid.

• SWE-1-mini: A small, extremely fast model that powers the Windsurf Tab passive experience for all users, free or paid.

These custom models are cheaper for Windsurf users than the frontier labs companies offerings and replace Cascade Base, Windsurf's former free (and underpowered) model.

The best way to understand these models in the context of Windsurf's previously confusing pricing model, since simplified. Windsurf had two kinds of tokens — and one was for "flow". Essentially, these are like task-completion tokens that are more valuable (and expensive) because they can complete sequences of actions rather than just token prediction.

SWE-1 has what Windsurf calls "flow awareness":

What is flow awareness? We built the Windsurf Editor in order to build a seamless intertwining between the comprehensive states of user and the AI. Anything that the AI does, the human should be able to observe and action on, and anything the human does, the AI should be able to observe and action on. We call this awareness of the shared timeline “flow awareness,” and thus why we have always referred to our collaborative agentic experience as “AI flows.”

Thus building models that are "flow aware" makes them much more powerful in the context of the IDE, where the agent or copilot can track what's happening across the entire application and codebase, continually updating a stack of edits or commands the developer makes.

As a result, Cascade can monitor your terminal commands, clipboard, current Cascade conversation, and in-IDE user searches to better understand the user's goal and therefore suggest much more relevant and contextual next steps.

Sound interesting? Sign up here!