benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
papers
articles
OpenAI
gpt-oss-20b
4 benchmarks
AIME
#5 of 117
98.7%
MMLU Pro
#14 of 55
85.3%
GPQA Diamond
#90 of 118
67.1%
SWE-bench Verified
#68 of 86
60.7%