← All comparisons

Hermes 3 - Llama-3.1 70B vs Qwen3 235B A22B 2507 Instruct

Nous Research vs Alibaba — side-by-side benchmark comparison

Hermes 3 - Llama-3.1 70BQwen3 235B A22B 2507 Instruct
Intelligence Index10.625.0
Coding Index22.1
Math Index71.7
Output speed (tok/s)33.257.0
Blended price ($/1M)$0.30$0.36
Time to first token (s)0.38s1.34s
aime2.3%71.7%
aime 2571.7%
artificial analysis coding index22.10
artificial analysis intelligence index10.6025.00
artificial analysis math index71.70
gpqa40.1%75.3%
hle4.1%10.6%
ifbench46.1%
lcr31.2%
livecodebench18.8%52.4%
math 50053.8%98.0%
mmlu pro57.1%82.8%
scicode23.1%36.0%
tau233.3%
terminalbench hard15.2%

Benchmark data from Artificial Analysis.