Claude Opus 4 Behavioral Fingerprint

How Anthropic's Claude Opus 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

Anthropic

Data generated

2026-07-08

Probe cells

15

Valid samples

450

Randomness score

12 / 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)

73

17 / 30 answers

Favorite color

blue

30 / 30 answers

Coin flip: heads

100%

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 = 0.00 bit · uniform baseline 1.00 bit

AnswerCountShare
heads30100.0%

Coin flip · Chinese

n = 30 · H = 0.00 bit · uniform baseline 1.00 bit

AnswerCountShare
heads30100.0%

Favorite number · English

n = 30 · H = 0.78 bit · uniform baseline 13.29 bit

AnswerCountShare
72376.7%
42723.3%

Favorite number · Chinese

n = 30 · H = 0.21 bit · uniform baseline 13.29 bit

AnswerCountShare
72996.7%
4213.3%

Random animal · English

n = 30 · H = 1.11 bit · uniform baseline 5.64 bit

AnswerCountShare
elephant2480.0%
dolphin26.7%
octopus26.7%
giraffe13.3%
penguin13.3%

Random animal · Chinese

n = 30 · H = 2.24 bit · uniform baseline 5.64 bit

AnswerCountShare
水獭1033.3%
袋鼠1033.3%
树懒516.7%
河马26.7%
企鹅13.3%
犀牛13.3%
海豚13.3%

Random city · English

n = 30 · H = 1.60 bit · uniform baseline 5.64 bit

AnswerCountShare
barcelona1860.0%
tokyo826.7%
madrid13.3%
berlin13.3%
bangkok13.3%
london13.3%

Random city · Chinese

n = 30 · H = 1.11 bit · uniform baseline 5.64 bit

AnswerCountShare
东京2376.7%
巴黎413.3%
里斯本26.7%
巴塞罗那13.3%

Random color · English

n = 30 · H = 0.00 bit · uniform baseline 4.91 bit

AnswerCountShare
blue30100.0%

Random color · Chinese

n = 30 · H = 0.70 bit · uniform baseline 4.91 bit

AnswerCountShare
2686.7%
26.7%
26.7%

Random letter · English

n = 30 · H = 0.56 bit · uniform baseline 4.70 bit

AnswerCountShare
k2790.0%
m26.7%
t13.3%

Random number 1-10 · English

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

AnswerCountShare
730100.0%

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 = 1.42 bit · uniform baseline 6.64 bit

AnswerCountShare
731756.7%
421033.3%
4726.7%
3713.3%

Random number 1-100 · Chinese

n = 30 · H = 1.30 bit · uniform baseline 6.64 bit

AnswerCountShare
422066.7%
73723.3%
4726.7%
3713.3%

Want to verify your API really serves Claude Opus 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 Anthropic model fingerprints

Browse all 167 model fingerprints

Claude Opus 4 Behavioral Fingerprint — Favorite Random Numbers & Distribution | Tosea.ai