Cognition Labs' AI software engineer. SWE-bench Verified 53.8%. First autonomous coding agent to demonstrate end-to-end software development.
Metrics
Info
ProviderCognition Labs
Organization that created the model
Licensecommercial
Software license type
Agent Typecoding
Agent specialization area
Base Model Requirementsproprietary
Required foundation models to operate
Other
Valuation2,000,000,000
Funding Total175,000,000
Swe Bench Verified53.8
Score Breakdown
adoption
15%85
Community usage, market traction, and ecosystem maturity
tool use
90
memory context
88
self reflection
86
planning reasoning
78
Compatibility(1)
Sources(2)
Developed by
Cognition LabsRelated Signals(2)
Claude Opus 4.5 First to Break 80% on SWE-bench Verified
AgentsJan 11
95Claude Opus 4.5 achieved 80.9% on SWE-bench Verified, becoming the first model to exceed the 80% threshold. This represents a 21x improvement from GPT-4's initial 3.8% score in October 2023, demonstrating rapid progress in AI coding capabilities.
Related:
Devin Pioneered Commercial AI Software Engineering
AgentsMar 12
88Devin emerged as the first commercial AI software engineer, achieving $2B valuation and 53.8% on SWE-bench Verified. As the pioneering commercial coding agent, it established the market category now contested by Claude Code, Cursor, and others.
Related: