Latest Models

Scores across benchmarks

Updated Jul 19, 2026, 10:00 UTC

aisocratic.org
#ModelLM ArenaSWE-benchARC-AGI-2
🥇
AnthropicClaude Opus 4.6
150475.6%69.2%($3.47)
🥈
ZhipuGLM 5
-72.8%22.8%($0.25)
🥉
OpenAIGPT 5.4
--83.3%($16.41)
4
OpenAIGPT 5.2
-72.8%72.9%($38.99)
5
GoogleGemini 3 Pro
148669.6%54.0%($30.57)
6
AnthropicClaude Opus 4.5
-76.8%37.6%($2.40)