gemini-4IMAGE
Gemini 4 Argon scores 26.3% on FrontierCode 2.0, ranking 11th below Claude Haiku 5.5
On Cognition's new FrontierCode 2.0 coding benchmark, Gemini 4 Argon scores 26.3% for 11th place, below Claude Haiku 5.5's 27.7% and far behind Opus 5.5's 45.5% and GPT-6 Astra's 42.8%.
Vipin
@vipinxtwtArchived 13 hours agoOriginally posted 2026.10.28
Vipin @vipinxtwtOriginally posted 2026.10.28
Gemini 4 Argon ranks 11th on cognition’s new FrontierCode 2.0 coding benchmark, scoring just 26.3%, even below claude haiku 5.5 at 27.7%. Claude opus 5.5 leads with 45.5%, followed by GPT-6 astra at 42.8%. The gap is bigger than I expected.
x.com/vipinxtwt/status/2115380804100468915Archived 2026.10.29 10:09