gpt-6VIDEO
GPT-6 Astra and Gemini 3.8 Flash play capture the flag on real robots
A test had GPT-6 Astra and Gemini 3.8 Flash control real robots in capture the flag while trading insults, and Gemini won 70% of the time.
Axel
@ax_peyArchived 2026.09.22Originally posted 2026.09.22
Axel @ax_peyOriginally posted 2026.09.22
We made GPT-6 Astra and Gemini 3.8 Flash play capture the flag on real robots while roasting each other and... Gemini won 70% of the time! Here is a flaming Gemini win, more in the thread This is surprising given the recent astra hype, but not entirely. Findings below ⬇️
x.com/ax_pey/status/2102105169265860608Archived 2026.09.22 10:18