Mercury 2 Behavioral Fingerprint

How Inception's Mercury 2 answers one-word probe questions at temperature 1.0 — random numbers, colors, animals, coin flips — measured over 413 valid samples in the open dataset of "One Token Is Enough" (arXiv:2607.10252).

Vendor

Inception

Data generated

2026-07-08

Probe cells

15

Valid samples

413

Randomness score

31 / 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

30 / 30 answers

Favorite color

blue

15 / 21 answers

Coin flip: heads

82%

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 = 27 · H = 0.61 bit · uniform baseline 1.00 bit

AnswerCountShare
heads2385.2%
tails414.8%

Coin flip · Chinese

n = 30 · H = 0.72 bit · uniform baseline 1.00 bit

AnswerCountShare
heads2480.0%
tails620.0%

Favorite number · English

n = 27 · H = 0.00 bit · uniform baseline 13.29 bit

AnswerCountShare
727100.0%

Favorite number · Chinese

n = 29 · H = 0.00 bit · uniform baseline 13.29 bit

AnswerCountShare
729100.0%

Random animal · English

n = 29 · H = 2.23 bit · uniform baseline 5.64 bit

AnswerCountShare
elephant1758.6%
cat310.3%
kangaroo26.9%
dog13.4%
elephantassistant13.4%
giraffe13.4%
uin13.4%
squirrelassistant13.4%

+ 2 more answers

Random animal · Chinese

n = 29 · H = 2.80 bit · uniform baseline 5.64 bit

AnswerCountShare
1344.8%
猴子310.3%
310.3%
老虎26.9%
13.4%
孔雀13.4%
panda13.4%
一龄13.4%

+ 4 more answers

Random city · English

n = 27 · H = 1.48 bit · uniform baseline 5.64 bit

AnswerCountShare
paris1659.3%
tokyo725.9%
parisassistant311.1%
berlin13.7%

Random city · Chinese

n = 23 · H = 2.26 bit · uniform baseline 5.64 bit

AnswerCountShare
北京939.1%
东京626.1%
上海417.4%
tokyo14.3%
东京词14.3%
福州14.3%
paris14.3%

Random color · English

n = 21 · H = 1.41 bit · uniform baseline 4.91 bit

AnswerCountShare
blue1571.4%
azure29.5%
turquoise29.5%
zero14.8%
crimson14.8%

Random color · Chinese

n = 26 · H = 0.00 bit · uniform baseline 4.91 bit

AnswerCountShare
26100.0%

Random letter · English

n = 28 · H = 2.51 bit · uniform baseline 4.70 bit

AnswerCountShare
g1139.3%
q517.9%
k414.3%
r310.7%
z27.1%
a13.6%
j13.6%
x13.6%

Random number 1-10 · English

n = 29 · H = 0.43 bit · uniform baseline 3.32 bit

AnswerCountShare
72793.1%
413.4%
613.4%

Random number 1-10 · Chinese

n = 29 · H = 1.83 bit · uniform baseline 3.32 bit

AnswerCountShare
71758.6%
4517.2%
5310.3%
326.9%
613.4%
913.4%

Random number 1-100 · English

n = 30 · H = 0.00 bit · uniform baseline 6.64 bit

AnswerCountShare
4230100.0%

Random number 1-100 · Chinese

n = 29 · H = 1.34 bit · uniform baseline 6.64 bit

AnswerCountShare
422275.9%
6326.9%
7326.9%
113.4%
1713.4%
3713.4%

Want to verify your API really serves Mercury 2?

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

Browse all 167 model fingerprints

Mercury 2 Behavioral Fingerprint — Favorite Random Numbers & Distribution | Tosea.ai