GLM-5.1 vs GPT-5.3 Codex — Who Actually Wins? Real Benchmark Data (2026)

head to head

GLM-5.1 vs GPT-5.3 Codex

4 shared benchmarks

1

wins

0

ties

3

wins

86.2%

GPQA Diamond

91.5%

58.4%

SWE-bench Pro

56.8%

77.8%

SWE-bench Verified

80.0%

69.0%

Terminal-Bench 2.0

77.3%