← All comparisons

Hermes 4 - Llama-3.1 405B (Reasoning) vs MiMo-V2.5

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

Hermes 4 - Llama-3.1 405B (Reasoning)MiMo-V2.5
Intelligence Index18.649.0
Coding Index16.042.1
Math Index69.7
Output speed (tok/s)38.692.5
Blended price ($/1M)$1.50$0.17
Time to first token (s)0.79s1.83s
aime
aime 2569.7%
artificial analysis coding index16.0042.10
artificial analysis intelligence index18.6049.00
artificial analysis math index69.70
gpqa72.7%84.9%
hle10.3%25.2%
ifbench32.7%67.1%
lcr20.7%62.7%
livecodebench68.6%
math 500
mmlu pro82.9%
scicode25.2%43.1%
tau222.2%90.6%
terminalbench hard11.4%41.7%

Benchmark data from Artificial Analysis.