claude 3.7 no thinking (diff) - 60.4%

claude 3.7 32k thinking tokens (diff) - 64.9%

GPT-4.1 (diff) - 52.9% (stat is from the blog post)

https://aider.chat/docs/leaderboards/