benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
papers
articles
head to head
Kimi K2.5
vs
Qwen 3.6 35B-A3B
11 shared benchmarks
9
wins
0
ties
2
wins
96.1%
AIME
92.7%
87.6%
GPQA Diamond
86.0%
24.37%
Humanity's Last Exam
21.4%
85.0%
LiveCodeBench
80.4%
87.1%
MMLU Pro
85.2%
78.5%
MMMU Pro
75.3%
89.33%
OmniDocBench
89.9%
73.0%
SWE-bench Multilingual
67.2%
50.7%
SWE-bench Pro
49.5%
76.8%
SWE-bench Verified
73.4%
50.8%
Terminal-Bench 2.0
51.5%