benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
papers
articles
OpenAI
GPT-5.6 Terra
9 benchmarks
GPQA Diamond
#15 of 118
92.9%
BrowseComp
#9 of 45
87.5%
Terminal-Bench 2.1
#6 of 27
87.4%
FrontierMath
#2 of 11
84.9%
MMMU Pro
#9 of 36
80.7%
DeepSWE
#4 of 22
69.6%
SWE-bench Pro
#8 of 43
63.4%
OSWorld 2.0
#3 of 10
50.2%
ARC-AGI 3
#2 of 8
0.8%