AI & Development
Claude Opus 4.5 Breaks 80% on SWE-Bench: First AI to Hit Human-Level Coding Milestone
Anthropic’s Claude Opus 4.5 became the first AI model to break 80% on SWE-bench Verified, ...