benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
papers
articles
head to head
DeepSeek V4 Pro
vs
GLM-5.1
7 shared benchmarks
4
wins
0
ties
3
wins
95.2%
AIME
95.3%
83.4%
BrowseComp
68.0%
90.1%
GPQA Diamond
86.2%
37.7%
Humanity's Last Exam
31.0%
55.4%
SWE-bench Pro
58.4%
80.6%
SWE-bench Verified
77.8%
67.9%
Terminal-Bench 2.0
69.0%