So 89.4% on Terminal Bench 2 but only 19.1% on Tbench 4. Opus 5 is 89.1%/51.8%. | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

leumon 14 days ago | parent | context | favorite | on: Gemini 3.8 Flash and 3.8 Flash Cyber

So 89.4% on Terminal Bench 2 but only 19.1% on Tbench 4. Opus 5 is 89.1%/51.8%.

aszen 13 days ago [–]

I was thinking the same, obvious suspicion is they benchmaxed it on older bench.

the_duke 13 days ago | [–]

Google has always done quite a lot of benchmaxing for Gemini.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact