← All comparisons

Hermes 4 - Llama-3.1 70B (Reasoning) vs EXAONE 4.0 32B (Non-reasoning)

Nous Research vs LG AI Research — side-by-side benchmark comparison

Hermes 4 - Llama-3.1 70B (Reasoning)EXAONE 4.0 32B (Non-reasoning)
Intelligence Index16.011.7
Coding Index14.49.4
Math Index68.739.3
Output speed (tok/s)92.80.0
Blended price ($/1M)$0.20$0.00
Time to first token (s)0.64s0.00s
aime47.0%
aime 2568.7%39.3%
artificial analysis coding index14.409.40
artificial analysis intelligence index16.0011.70
artificial analysis math index68.7039.30
gpqa69.9%62.8%
hle7.9%4.9%
ifbench31.3%33.5%
lcr6.7%8.0%
livecodebench65.3%47.2%
math 50093.9%
mmlu pro81.1%76.8%
scicode34.1%25.2%
tau222.5%4.1%
terminalbench hard4.5%1.5%

Benchmark data from Artificial Analysis.