← All comparisons

Hermes 4 - Llama-3.1 405B (Reasoning) vs Qwen3 Next 80B A3B (Reasoning)

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

Hermes 4 - Llama-3.1 405B (Reasoning)Qwen3 Next 80B A3B (Reasoning)
Intelligence Index18.626.7
Coding Index16.019.5
Math Index69.784.3
Output speed (tok/s)38.6149.4
Blended price ($/1M)$1.50$1.88
Time to first token (s)0.79s1.23s
aime
aime 2569.7%84.3%
artificial analysis coding index16.0019.50
artificial analysis intelligence index18.6026.70
artificial analysis math index69.7084.30
gpqa72.7%75.9%
hle10.3%11.7%
ifbench32.7%60.7%
lcr20.7%60.3%
livecodebench68.6%78.4%
math 500
mmlu pro82.9%82.4%
scicode25.2%38.8%
tau222.2%41.5%
terminalbench hard11.4%9.8%

Benchmark data from Artificial Analysis.