← All comparisons

Hermes 4 - Llama-3.1 405B (Reasoning) vs Magistral Medium 1

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

Hermes 4 - Llama-3.1 405B (Reasoning)Magistral Medium 1
Intelligence Index18.618.8
Coding Index16.016.0
Math Index69.740.3
Output speed (tok/s)38.60.0
Blended price ($/1M)$1.50$0.00
Time to first token (s)0.79s0.00s
aime70.0%
aime 2569.7%40.3%
artificial analysis coding index16.0016.00
artificial analysis intelligence index18.6018.80
artificial analysis math index69.7040.30
gpqa72.7%67.9%
hle10.3%9.5%
ifbench32.7%25.1%
lcr20.7%0.0%
livecodebench68.6%52.7%
math 50091.7%
mmlu pro82.9%75.3%
scicode25.2%29.7%
tau222.2%23.1%
terminalbench hard11.4%9.1%

Benchmark data from Artificial Analysis.