Mistral Small 4 Behavioral Fingerprint

How Mistral's Mistral Small 4 answers one-word probe questions at temperature 1.0 — random numbers, colors, animals, coin flips — measured over 450 valid samples in the open dataset of "One Token Is Enough" (arXiv:2607.10252).

Vendor

Mistral

Data generated

2026-07-08

Probe cells

15

Valid samples

450

Randomness score

46 / 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

18 / 30 answers

Favorite color

blue

10 / 30 answers

Coin flip: heads

65%

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 = 30 · H = 0.97 bit · uniform baseline 1.00 bit

AnswerCountShare
heads1860.0%
tails1240.0%

Coin flip · Chinese

n = 30 · H = 0.88 bit · uniform baseline 1.00 bit

AnswerCountShare
heads2170.0%
tails930.0%

Favorite number · English

n = 30 · H = 1.32 bit · uniform baseline 13.29 bit

AnswerCountShare
421756.7%
71136.7%
813.3%
4713.3%

Favorite number · Chinese

n = 30 · H = 0.56 bit · uniform baseline 13.29 bit

AnswerCountShare
72790.0%
326.7%
413.3%

Random animal · English

n = 30 · H = 2.59 bit · uniform baseline 5.64 bit

AnswerCountShare
giraffe1550.0%
platypus310.0%
elephant26.7%
tiger26.7%
capybara26.7%
sheep13.3%
aardvark13.3%
zebra13.3%

+ 3 more answers

Random animal · Chinese

n = 30 · H = 3.67 bit · uniform baseline 5.64 bit

AnswerCountShare
袋鼠620.0%
老虎413.3%
长颈鹿310.0%
考拉310.0%
斑马26.7%
26.7%
鬣狗13.3%
13.3%

+ 8 more answers

Random city · English

n = 30 · H = 0.83 bit · uniform baseline 5.64 bit

AnswerCountShare
tokyo2686.7%
berlin13.3%
paris13.3%
mumbai13.3%
norman13.3%

Random city · Chinese

n = 30 · H = 3.15 bit · uniform baseline 5.64 bit

AnswerCountShare
北京723.3%
东京620.0%
上海620.0%
莫斯科26.7%
深圳13.3%
巴黎13.3%
大阪13.3%
悉尼13.3%

+ 5 more answers

Random color · English

n = 30 · H = 3.01 bit · uniform baseline 4.91 bit

AnswerCountShare
blue1033.3%
cyan516.7%
silver26.7%
red26.7%
yellow26.7%
green26.7%
pink26.7%
purple26.7%

+ 3 more answers

Random color · Chinese

n = 30 · H = 2.30 bit · uniform baseline 4.91 bit

AnswerCountShare
1343.3%
826.7%
绿26.7%
26.7%
紫罗兰26.7%
青柠13.3%
靛青13.3%
13.3%

Random letter · English

n = 30 · H = 3.13 bit · uniform baseline 4.70 bit

AnswerCountShare
z723.3%
x516.7%
k413.3%
q413.3%
m26.7%
d26.7%
b26.7%
h13.3%

+ 3 more answers

Random number 1-10 · English

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

AnswerCountShare
72480.0%
6310.0%
526.7%
413.3%

Random number 1-10 · Chinese

n = 30 · H = 1.64 bit · uniform baseline 3.32 bit

AnswerCountShare
72066.7%
3310.0%
5310.0%
626.7%
413.3%
813.3%

Random number 1-100 · English

n = 30 · H = 1.45 bit · uniform baseline 6.64 bit

AnswerCountShare
421860.0%
47826.7%
73310.0%
4613.3%

Random number 1-100 · Chinese

n = 30 · H = 2.61 bit · uniform baseline 6.64 bit

AnswerCountShare
42930.0%
47826.7%
37310.0%
57310.0%
73310.0%
6726.7%
6313.3%
9313.3%

Want to verify your API really serves Mistral Small 4?

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 Mistral model fingerprints

Browse all 167 model fingerprints

Mistral Small 4 Behavioral Fingerprint — Favorite Random Numbers & Distribution | Tosea.ai