← All comparisons

Hermes 4 - Llama-3.1 405B (Non-reasoning) vs DeepSeek R1 Distill Llama 8B

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

Hermes 4 - Llama-3.1 405B (Non-reasoning)DeepSeek R1 Distill Llama 8B
Intelligence Index17.612.1
Coding Index18.1
Math Index15.341.3
Output speed (tok/s)40.80.0
Blended price ($/1M)$1.50$0.00
Time to first token (s)0.73s0.00s
aime33.3%
aime 2515.3%41.3%
artificial analysis coding index18.10
artificial analysis intelligence index17.6012.10
artificial analysis math index15.3041.30
gpqa53.6%30.2%
hle4.2%4.2%
ifbench34.8%17.6%
lcr20.0%0.0%
livecodebench54.6%23.3%
math 50085.3%
mmlu pro72.9%54.3%
scicode34.6%11.9%
tau226.6%
terminalbench hard9.8%

Benchmark data from Artificial Analysis.