gemini-4IMAGE
Gemini 4 Argon competes with GPT-6.1 Sol on HLE-Diamond at ~5x output tokens
On HLE-Diamond, Gemini 4 Argon is competitive with GPT-6.1 Sol and Opus 5.5 but uses about five times more output tokens than GPT-6 models.
Long Phan
@longphann_Archived 21 days agoOriginally posted 2026.10.15

Long Phan @longphann_Originally posted 2026.10.15
Gemini 4 Argon and Muse Wonder on HLE-Diamond. Gemini 4 Argon competes with GPT-6.1 Sol and Opus 5.5, but uses ~5x more output tokens than GPT-6 models.
x.com/longphann_/status/2110408432897675593Archived 2026.10.15 11:35