← All comparisons

Hermes 4 - Llama-3.1 70B (Non-reasoning) vs Mi:dm K 2.5 Pro Preview

Nous Research vs Korea Telecom — side-by-side benchmark comparison

Hermes 4 - Llama-3.1 70B (Non-reasoning)Mi:dm K 2.5 Pro Preview
Intelligence Index12.6
Coding Index9.211.9
Math Index11.378.7
Output speed (tok/s)94.30.0
Blended price ($/1M)$0.20$0.00
Time to first token (s)0.61s0.00s
aime
aime 2511.3%78.7%
artificial analysis coding index9.2011.90
artificial analysis intelligence index12.60
artificial analysis math index11.3078.70
gpqa49.1%72.2%
hle3.6%8.8%
ifbench29.0%45.6%
lcr2.0%11.0%
livecodebench26.9%57.6%
math 500
mmlu pro66.4%81.3%
scicode27.7%29.7%
tau221.6%49.4%
terminalbench hard0.0%3.0%

Benchmark data from Artificial Analysis.