LLMs Conquered: A Graveyard of AI Benchmarks

2025-01-06

Killedbyllm.com is a fascinating website documenting the rapid progress of Large Language Models (LLMs). It lists numerous benchmarks, from early reading comprehension tests to complex mathematical reasoning challenges, that have been surpassed by models like GPT-4 and LLaMa. The site serves as a testament to the breakneck speed of AI advancement, showcasing how previously insurmountable challenges have fallen to LLMs and prompting reflection on the future of AI.