← All comparisons

Hermes 4 - Llama-3.1 405B (Reasoning) vs Mistral Medium

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

Hermes 4 - Llama-3.1 405B (Reasoning)Mistral Medium
Intelligence Index18.69.0
Coding Index16.0
Math Index69.7
Output speed (tok/s)38.667.5
Blended price ($/1M)$1.50$4.09
Time to first token (s)0.79s0.56s
aime3.7%
aime 2569.7%
artificial analysis coding index16.00
artificial analysis intelligence index18.609.00
artificial analysis math index69.70
gpqa72.7%34.9%
hle10.3%3.4%
ifbench32.7%
lcr20.7%
livecodebench68.6%9.9%
math 50040.5%
mmlu pro82.9%49.1%
scicode25.2%11.8%
tau222.2%
terminalbench hard11.4%

Benchmark data from Artificial Analysis.