← All comparisons

Hermes 4 - Llama-3.1 405B (Non-reasoning) vs Qwen3.5 27B (Non-reasoning)

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

Hermes 4 - Llama-3.1 405B (Non-reasoning)Qwen3.5 27B (Non-reasoning)
Intelligence Index17.637.2
Coding Index18.133.4
Math Index15.3
Output speed (tok/s)40.895.3
Blended price ($/1M)$1.50$0.88
Time to first token (s)0.73s1.40s
aime
aime 2515.3%
artificial analysis coding index18.1033.40
artificial analysis intelligence index17.6037.20
artificial analysis math index15.30
gpqa53.6%84.2%
hle4.2%13.2%
ifbench34.8%46.9%
lcr20.0%55.7%
livecodebench54.6%
math 500
mmlu pro72.9%
scicode34.6%36.7%
tau226.6%87.1%
terminalbench hard9.8%31.8%

Benchmark data from Artificial Analysis.