Claude Opus 4.6 vs GPT-5.4 — Who Actually Wins? Real Benchmark Data (2026)

head to head

Claude Opus 4.6 vs GPT-5.4

9 shared benchmarks

4

wins

1

ties

4

wins

85.0%

ARC-AGI 2

73.3%

0.25%

ARC-AGI 3

0.26%

84.0%

BrowseComp

82.7%

91.3%

GPQA Diamond

92.8%

53.1%

Humanity's Last Exam

36.24%

72.7%

OSWorld

75.0%

81.42%

SWE-bench Verified

77.2%

81.8%

Terminal-Bench 2.0

81.8%

42.3%

USAMO 2026

95.2%