Claude Opus 4.7 vs Claude Opus 4.8 — Who Actually Wins? Real Benchmark Data (2026)

head to head

Claude Opus 4.7 vs Claude Opus 4.8

9 shared benchmarks

3

wins

0

ties

6

wins

79.3%

BrowseComp

84.3%

82.1%

CharXiv

80.5%

94.2%

GPQA Diamond

91.0%

46.9%

Humanity's Last Exam

49.8%

79.1%

MCP-Atlas

77.8%

75.2%

MMMU Pro

78.9%

48.9%

OSWorld 2.0

54.8%

64.3%

SWE-bench Pro

69.2%

66.1%

Terminal-Bench 2.1

84.6%