benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
papers
articles
head to head
Gemini 3.1 Pro
vs
Kimi K2.6
11 shared benchmarks
8
wins
0
ties
3
wins
98.13%
AIME
96.4%
85.9%
BrowseComp
83.2%
83.3%
CharXiv
80.4%
94.3%
GPQA Diamond
90.5%
44.4%
Humanity's Last Exam
34.7%
80.5%
MMMU Pro
79.4%
72.7%
OSWorld
73.1%
54.2%
SWE-bench Pro
58.6%
80.6%
SWE-bench Verified
80.2%
68.5%
Terminal-Bench 2.0
66.7%
70.3%
Terminal-Bench 2.1
71.3%