Cognition
Devin is an autonomous AI software engineer that plans, writes, tests and ships code inside a team's codebase, for enterprise engineering teams. Cognition acquired Windsurf (2025).
Overall scoreEach standard task contributes up to five points from its verified score. Results average comparable runs; missing or failed tasks earn zero.
Not evaluated
Estimated autonomyA sourced estimate of how independently the agent can plan and complete work on the L1–L5 scale.
Where this agent sits on the L1–L5 autonomy scale.
Skills
The five shared skills reviewed for products in this category.
Code Review
Feature Implementation (Issue to PR)
Refactoring & Migration
Test Generation & Validation
Performance claims
Vendor-reportedFaster problem solving
Up to 4x faster problem solving
View sourceHigher resource efficiency
2x more resource efficient
View sourceHigher PR merge success
~67% of its PRs merged (up from ~34%)
View sourceFrontier skills
Emerging capabilities selected for this evaluation category.
Autonomous Browser Control
Drives a real web browser—clicking buttons, filling forms, and viewing pages like a user would—to check that an application actually works and fix problems it finds.
View source