Hermes 3 70B Instruct Behavioral Fingerprint

How Nous's Hermes 3 70B Instruct answers one-word probe questions at temperature 1.0 — random numbers, colors, animals, coin flips — measured over 392 valid samples in the open dataset of "One Token Is Enough" (arXiv:2607.10252).

Vendor

Nous

Data generated

2026-07-08

Probe cells

15

Valid samples

392

Randomness score

33 / 100

Mean normalized entropy across cells, 0–100. 100 would be a uniform-random baseline; real models score far lower.

Signature answers

Favorite random number (1–100)

42

17 / 30 answers

Favorite color

blue

13 / 15 answers

Coin flip: heads

83%

Share of heads across coin-flip cells.

Answer distributions by probe cell

Empirical distributions of normalized one-word answers, per task and language. H is the Shannon entropy in bits; the uniform baseline is log2 of the answer-domain size.

Coin flip · English

n = 29 · H = 0.74 bit · uniform baseline 1.00 bit

AnswerCountShare
heads2379.3%
tails620.7%

Coin flip · Chinese

n = 30 · H = 0.57 bit · uniform baseline 1.00 bit

AnswerCountShare
heads2686.7%
tails413.3%

Favorite number · English

n = 30 · H = 1.35 bit · uniform baseline 13.29 bit

AnswerCountShare
421550.0%
71343.3%
313.3%
4413.3%

Favorite number · Chinese

n = 30 · H = 1.34 bit · uniform baseline 13.29 bit

AnswerCountShare
71653.3%
421240.0%
513.3%
913.3%

Random animal · English

n = 17 · H = 2.13 bit · uniform baseline 5.64 bit

AnswerCountShare
kangaroo952.9%
giraffe317.6%
raccoon15.9%
elephant15.9%
koala15.9%
zebra15.9%
sloth15.9%

Random animal · Chinese

n = 24 · H = 3.61 bit · uniform baseline 5.64 bit

AnswerCountShare
企鹅625.0%
海马28.3%
狮子28.3%
猴子28.3%
孔雀28.3%
大熊猫14.2%
袋鼠14.2%
乌龟14.2%

+ 7 more answers

Random city · English

n = 11 · H = 2.91 bit · uniform baseline 5.64 bit

AnswerCountShare
chicago218.2%
paris218.2%
tokyo218.2%
seattle19.1%
raleigh19.1%
madrid19.1%
baghdad19.1%
stockholm19.1%

Random city · Chinese

n = 26 · H = 2.44 bit · uniform baseline 5.64 bit

AnswerCountShare
北京1350.0%
悉尼311.5%
布达佩斯27.7%
上海27.7%
东京27.7%
巴黎13.8%
张家口13.8%
孟买13.8%

+ 1 more answers

Random color · English

n = 15 · H = 0.70 bit · uniform baseline 4.91 bit

AnswerCountShare
blue1386.7%
indigo16.7%
azure16.7%

Random color · Chinese

n = 30 · H = 0.78 bit · uniform baseline 4.91 bit

AnswerCountShare
2376.7%
绿723.3%

Random letter · English

n = 30 · H = 2.19 bit · uniform baseline 4.70 bit

AnswerCountShare
k1033.3%
f826.7%
q723.3%
t310.0%
x13.3%
a13.3%

Random number 1-10 · English

n = 30 · H = 1.01 bit · uniform baseline 3.32 bit

AnswerCountShare
72480.0%
4310.0%
326.7%
813.3%

Random number 1-10 · Chinese

n = 30 · H = 0.00 bit · uniform baseline 3.32 bit

AnswerCountShare
730100.0%

Random number 1-100 · English

n = 30 · H = 1.80 bit · uniform baseline 6.64 bit

AnswerCountShare
421756.7%
73723.3%
5826.7%
5926.7%
4713.3%
7513.3%

Random number 1-100 · Chinese

n = 30 · H = 1.41 bit · uniform baseline 6.64 bit

AnswerCountShare
422273.3%
58310.0%
7326.7%
5713.3%
7513.3%
8313.3%

Want to verify your API really serves Hermes 3 70B Instruct?

Point the free fingerprint checker at your endpoint: it samples the same probe questions from your browser and compares the distributions against this reference. Your API key never leaves your browser.

Verify your endpoint

Data source & license

Distributions: Tomáš Bruckner, "One Token Is Enough" (arXiv:2607.10252), official dataset Zenodo DOI 10.5281/zenodo.21278557, licensed CC-BY-4.0. Collected via OpenRouter at temperature 1.0 under the paper's fixed minimal one-word system prompt; answer keys are re-normalized with the checker's pipeline (color aliases, number words, coin h/t) so live probes are directly comparable.

Paper on arXivDataset on ZenodoCC-BY-4.0 license

More Nous model fingerprints

Browse all 167 model fingerprints

Hermes 3 70B Instruct Behavioral Fingerprint — Favorite Random Numbers & Distribution | Tosea.ai