Coding agent benchmark: Gemini 4 Argon lands fifth behind GPT-6 Astra

On a coding-agent benchmark spanning 11 unfamiliar ML and data-analysis tasks, Gemini 4 Argon places fifth at 0.202 while GPT-6 Astra leads at 0.422.

anya

@anya_yfa7Archived 17 days agoOriginally posted 2026.10.18

Coding agent benchmark: Gemini 4 Argon lands fifth behind GPT-6 Astra (image 1 of 2)

Coding agent benchmark: Gemini 4 Argon lands fifth behind GPT-6 Astra (image 2 of 2)

anya @anya_yfa7Originally posted 2026.10.18

Open original on X

GPT 6 Astra still leading 👀 → GPT-6 Astra: 0.422 → GPT-6.1 Sol: 0.383 → Claude Opus 5.5: 0.312 Gemini 4 Argon enters at #5 with 0.202, narrowly ahead of Sonnet 5.5 This benchmark tests coding agents across 11 unfamiliar ML and data analysis tasks Argon makes the top five but the gap to the leaders is clear here

x.com/anya_yfa7/status/2111785209670897998Archived 2026.10.19 09:58

Open original on X