@johnowhitaker 2024-08-02 ♥3 ↻1 original ↗
For the first ~5k games I used GPT-4o (not mini) and the win stats were:
Flash 1.5 1314
GPT-4o 858
Haiku 693

Good work Google Gemini Flash! Part of this is speed-related (first guess right wins) so it'll be interesting to re-run offline without the time part.

author:johnowhitaker kind:tweet model:gpt-4o on:gemini-1-5-flash year:2024

cited on: gemini-1-5-flash

Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.