Issue 001 tracked Anthropic’s slope toward saturating SWE-bench Verified; a day later the reported frontier is a different lab on a harder variant.
Development
GPT-5.2 Thinking sets SOTA on SWE-Bench Pro at 55.6%
Claim
GPT-5.2 Thinking set a new state of the art on SWE-Bench Pro at 55.6%.
Reported metric: 55.6%
Reporting date: (not necessarily the event date)
Sources
Attribution and reporting linked by this record; links do not imply independent verification.
- Welcome to December 12, 2025 (theinnermostloop.substack.com)
Cite or correct this record
Suggest a correction — opens a GitHub draft for you to review and submit.