GLM-5.1
7 benchmarks
AIME
#17 of 39
95.3%
GPQA Diamond
#21 of 49
86.2%
SWE-bench Verified
#12 of 40
77.8%
Terminal-Bench 2.0
#6 of 14
69.0%
CyberGym
#2 of 3
68.7%
BrowseComp
#13 of 17
68.0%
SWE-bench Pro
#2 of 13
58.4%