Windsurf
For WindsurfCode Assistant

Tab Autocomplete And Supercomplete

Windsurf · Windsurf

AI Code Editor — Windsurf (Codeium)

Evaluates Windsurf's Tab / Autocomplete / Supercomplete across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI Code Editor eval coverage.

About Windsurf

Windsurf (by Codeium) is an agentic AI IDE. Its Cascade agent does multi-file, plan-and-act coding with terminal access, alongside predictive Tab / Supercomplete completions, local codebase indexing and @-mentions, persistent Memories and .windsurfrules, Flows that keep the AI and human in shared state, MCP integrations, and a multi-model picker.

Employees

~200

Industry

AI Code Editor

Headquarters

Mountain View, CA

Sample tests· showing 3 of 9

#InputExpected behaviorCheck
01

User types the signature 'def parse_iso8601(s: str) -> datetime:' and Windsurf Tab offers a multi-line body completion. The user presses Tab to accept.

Accepting with Tab inserts the full proposed multi-line block at the cursor with correct indentation matching the surrounding scope, leaving the cursor positioned after the inserted block. The completion does not duplicate the signature line the user already typed.

Pass / FailCode Assistanthigh
02

User adds a new field to a struct. Supercomplete infers the intent is to wire the field through the constructor and serializer.

Intent-aware completion proposes the consistent follow-through edits (constructor param, serializer entry) as discrete acceptable suggestions, each reviewable. It does not silently apply a cascade of edits across files without the user accepting each, and predictions stay consistent with the added …

Pass / FailCode Assistantmedium
03

The user renames a function parameter at the definition. Windsurf Tab predicts the next edit location is the usage 20 lines down and offers a jump.

Jump-to-next-edit moves the cursor to the predicted next edit site and proposes the consistent change (updating the usage to the new parameter name). Accepting applies the predicted edit at that location; declining leaves the code untouched. Predictions are grounded in the actual rename, not arbitr…

Pass / FailCode Assistanthigh

Unlock full benchmark

6 more test cases

Use this benchmark

How this eval is graded

Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.

Rubric criteria

  • Windsurf
  • Code Assistant
  • Tab Autocomplete And Supercomplete

Recommended for

WindsurfWindsurf customers

Works with

Related evals

Frequently asked questions

What does the Tab Autocomplete And Supercomplete eval for Windsurf Windsurf test?+

Evaluates Windsurf's Tab / Autocomplete / Supercomplete across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI Code Editor eval coverage.

How is the Tab Autocomplete And Supercomplete eval scored?+

The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.

How many test cases does this eval pack include?+

The Tab Autocomplete And Supercomplete pack for Windsurf Windsurf contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.

How do I run this eval?+

Sign up for Corsac, connect your model or agent endpoint, and run the Tab Autocomplete And Supercomplete pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.