← All comparisons

Olmo 3 32B Think vs Hermes 3 - Llama-3.1 70B

Allen Institute for AI vs Nous Research — side-by-side benchmark comparison

Olmo 3 32B ThinkHermes 3 - Llama-3.1 70B
Intelligence Index12.110.6
Coding Index10.5—
Math Index73.7—
Output speed (tok/s)0.033.2
Blended price ($/1M)$0.00$0.30
Time to first token (s)0.00s0.38s
aime—2.3%
aime 2573.7%—
artificial analysis coding index10.50—
artificial analysis intelligence index12.1010.60
artificial analysis math index73.70—
gpqa61.0%40.1%
hle5.9%4.1%
ifbench49.1%—
lcr0.0%—
livecodebench67.2%18.8%
math 500—53.8%
mmlu pro75.9%57.1%
scicode28.6%23.1%
tau20.0%—
terminalbench hard1.5%—

Benchmark data from Artificial Analysis.