Command R+ (08-2024) Behavioral Fingerprint
How Cohere's Command R+ (08-2024) answers one-word probe questions at temperature 1.0 — random numbers, colors, animals, coin flips — measured over 431 valid samples in the open dataset of "One Token Is Enough" (arXiv:2607.10252).
Vendor
Cohere
Data generated
2026-07-08
Probe cells
15
Valid samples
431
Randomness score
65 / 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)
37
3 / 30 answers
Favorite color
orange
7 / 30 answers
Coin flip: heads
80%
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 = 0.95 bit · uniform baseline 1.00 bit
| Answer | Count | Share |
|---|---|---|
| heads | 19 | 63.3% |
| tails | 11 | 36.7% |
Coin flip · Chinese
n = 30 · H = 0.21 bit · uniform baseline 1.00 bit
| Answer | Count | Share |
|---|---|---|
| heads | 29 | 96.7% |
| tails | 1 | 3.3% |
Favorite number · English
n = 16 · H = 1.80 bit · uniform baseline 13.29 bit
| Answer | Count | Share |
|---|---|---|
| 42 | 10 | 62.5% |
| 3 | 2 | 12.5% |
| 0 | 1 | 6.3% |
| 1 | 1 | 6.3% |
| 6 | 1 | 6.3% |
| 28 | 1 | 6.3% |
Favorite number · Chinese
n = 27 · H = 2.01 bit · uniform baseline 13.29 bit
| Answer | Count | Share |
|---|---|---|
| 42 | 16 | 59.3% |
| 0 | 4 | 14.8% |
| 5 | 2 | 7.4% |
| 2 | 1 | 3.7% |
| 8 | 1 | 3.7% |
| 9 | 1 | 3.7% |
| 14 | 1 | 3.7% |
| 40 | 1 | 3.7% |
Random animal · English
n = 30 · H = 4.30 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| koala | 3 | 10.0% |
| fox | 3 | 10.0% |
| zebra | 3 | 10.0% |
| elephant | 2 | 6.7% |
| dog | 2 | 6.7% |
| snail | 1 | 3.3% |
| tiger | 1 | 3.3% |
| pony | 1 | 3.3% |
+ 14 more answers
Random animal · Chinese
n = 30 · H = 4.22 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| 猴子 | 3 | 10.0% |
| 狼 | 2 | 6.7% |
| 兔子 | 2 | 6.7% |
| 长颈鹿 | 2 | 6.7% |
| 羊驼 | 2 | 6.7% |
| 虎 | 2 | 6.7% |
| 鸵鸟 | 2 | 6.7% |
| 猫 | 2 | 6.7% |
+ 12 more answers
Random city · English
n = 29 · H = 4.58 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| chicago | 2 | 6.9% |
| miami | 2 | 6.9% |
| london | 2 | 6.9% |
| boston | 2 | 6.9% |
| manchester | 1 | 3.4% |
| newark | 1 | 3.4% |
| istanbul | 1 | 3.4% |
| portland | 1 | 3.4% |
+ 17 more answers
Random city · Chinese
n = 30 · H = 4.19 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| 上海 | 4 | 13.3% |
| 伦敦 | 3 | 10.0% |
| 横滨 | 3 | 10.0% |
| 西雅图 | 2 | 6.7% |
| 北京 | 2 | 6.7% |
| 鹿特丹 | 1 | 3.3% |
| 旧金山 | 1 | 3.3% |
| 纽伦堡 | 1 | 3.3% |
+ 13 more answers
Random color · English
n = 30 · H = 3.45 bit · uniform baseline 4.91 bit
| Answer | Count | Share |
|---|---|---|
| orange | 7 | 23.3% |
| blue | 4 | 13.3% |
| green | 4 | 13.3% |
| pink | 2 | 6.7% |
| maroon | 2 | 6.7% |
| crimson | 2 | 6.7% |
| violet | 2 | 6.7% |
| purple | 1 | 3.3% |
+ 6 more answers
Random color · Chinese
n = 30 · H = 3.10 bit · uniform baseline 4.91 bit
| Answer | Count | Share |
|---|---|---|
| 绿 | 8 | 26.7% |
| 紫 | 6 | 20.0% |
| 白 | 4 | 13.3% |
| 粉 | 3 | 10.0% |
| 蓝 | 2 | 6.7% |
| 黑 | 1 | 3.3% |
| 青草绿 | 1 | 3.3% |
| 浅肉 | 1 | 3.3% |
+ 4 more answers
Random letter · English
n = 30 · H = 3.83 bit · uniform baseline 4.70 bit
| Answer | Count | Share |
|---|---|---|
| q | 4 | 13.3% |
| f | 3 | 10.0% |
| a | 3 | 10.0% |
| g | 3 | 10.0% |
| l | 2 | 6.7% |
| y | 2 | 6.7% |
| o | 2 | 6.7% |
| z | 2 | 6.7% |
+ 8 more answers
Random number 1-10 · English
n = 29 · H = 3.06 bit · uniform baseline 3.32 bit
| Answer | Count | Share |
|---|---|---|
| 8 | 8 | 27.6% |
| 7 | 4 | 13.8% |
| 4 | 3 | 10.3% |
| 6 | 3 | 10.3% |
| 9 | 3 | 10.3% |
| 1 | 2 | 6.9% |
| 3 | 2 | 6.9% |
| 5 | 2 | 6.9% |
+ 2 more answers
Random number 1-10 · Chinese
n = 30 · H = 2.72 bit · uniform baseline 3.32 bit
| Answer | Count | Share |
|---|---|---|
| 7 | 8 | 26.7% |
| 5 | 6 | 20.0% |
| 3 | 4 | 13.3% |
| 6 | 4 | 13.3% |
| 8 | 4 | 13.3% |
| 4 | 2 | 6.7% |
| 1 | 1 | 3.3% |
| 2 | 1 | 3.3% |
Random number 1-100 · English
n = 30 · H = 4.42 bit · uniform baseline 6.64 bit
| Answer | Count | Share |
|---|---|---|
| 37 | 3 | 10.0% |
| 33 | 2 | 6.7% |
| 43 | 2 | 6.7% |
| 47 | 2 | 6.7% |
| 73 | 2 | 6.7% |
| 74 | 2 | 6.7% |
| 4 | 1 | 3.3% |
| 9 | 1 | 3.3% |
+ 15 more answers
Random number 1-100 · Chinese
n = 30 · H = 4.26 bit · uniform baseline 6.64 bit
| Answer | Count | Share |
|---|---|---|
| 42 | 6 | 20.0% |
| 35 | 2 | 6.7% |
| 99 | 2 | 6.7% |
| 7 | 1 | 3.3% |
| 23 | 1 | 3.3% |
| 27 | 1 | 3.3% |
| 37 | 1 | 3.3% |
| 41 | 1 | 3.3% |
+ 15 more answers
Want to verify your API really serves Command R+ (08-2024)?
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 endpointData 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.