← All comparisons

DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) vs GPT-5 (low)

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

DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)GPT-5 (low)
Intelligence Index7.639.2
Coding Index30.7
Math Index83.0
Output speed (tok/s)0.088.9
Blended price ($/1M)$0.00$3.44
Time to first token (s)0.00s8.32s
aime0.0%83.0%
aime 2583.0%
artificial analysis coding index30.70
artificial analysis intelligence index7.6039.20
artificial analysis math index83.00
gpqa27.0%80.8%
hle4.3%18.4%
ifbench66.6%
lcr58.7%
livecodebench8.5%76.3%
math 50021.8%98.7%
mmlu pro36.5%86.0%
scicode9.1%39.1%
tau284.2%
terminalbench hard26.5%

Benchmark data from Artificial Analysis.