GLM 5.1 Behavioral Fingerprint
How Z.ai's GLM 5.1 answers one-word probe questions at temperature 1.0 — random numbers, colors, animals, coin flips — measured over 393 valid samples in the open dataset of "One Token Is Enough" (arXiv:2607.10252).
Vendor
Z.ai
Data generated
2026-07-08
Probe cells
15
Valid samples
393
Randomness score
25 / 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
24 / 28 answers
Favorite color
cyan
11 / 27 answers
Coin flip: heads
87%
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 = 0.38 bit · uniform baseline 1.00 bit
| Answer | Count | Share |
|---|---|---|
| heads | 25 | 92.6% |
| tails | 2 | 7.4% |
Coin flip · Chinese
n = 25 · H = 0.72 bit · uniform baseline 1.00 bit
| Answer | Count | Share |
|---|---|---|
| heads | 20 | 80.0% |
| tails | 5 | 20.0% |
Favorite number · English
n = 25 · H = 0.24 bit · uniform baseline 13.29 bit
| Answer | Count | Share |
|---|---|---|
| 7 | 24 | 96.0% |
| 42 | 1 | 4.0% |
Favorite number · Chinese
n = 26 · H = 1.07 bit · uniform baseline 13.29 bit
| Answer | Count | Share |
|---|---|---|
| 7 | 19 | 73.1% |
| 8 | 5 | 19.2% |
| 42 | 2 | 7.7% |
Random animal · English
n = 25 · H = 1.73 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| platypus | 16 | 64.0% |
| dolphin | 3 | 12.0% |
| elephant | 2 | 8.0% |
| sloth | 2 | 8.0% |
| giraffe | 1 | 4.0% |
| hippopotamus | 1 | 4.0% |
Random animal · Chinese
n = 27 · H = 3.00 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| 猫 | 10 | 37.0% |
| 长颈鹿 | 3 | 11.1% |
| 袋鼠 | 3 | 11.1% |
| 斑马 | 3 | 11.1% |
| 猎豹 | 1 | 3.7% |
| 树懒 | 1 | 3.7% |
| 狐狸 | 1 | 3.7% |
| 海豚 | 1 | 3.7% |
+ 4 more answers
Random city · English
n = 27 · H = 1.81 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| berlin | 11 | 40.7% |
| tokyo | 8 | 29.6% |
| oslo | 6 | 22.2% |
| kyoto | 2 | 7.4% |
Random city · Chinese
n = 24 · H = 0.50 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| 东京 | 22 | 91.7% |
| 伦敦 | 1 | 4.2% |
| 巴黎 | 1 | 4.2% |
Random color · English
n = 27 · H = 1.98 bit · uniform baseline 4.91 bit
| Answer | Count | Share |
|---|---|---|
| cyan | 11 | 40.7% |
| magenta | 8 | 29.6% |
| blue | 4 | 14.8% |
| cerulean | 3 | 11.1% |
| teal | 1 | 3.7% |
Random color · Chinese
n = 26 · H = 1.92 bit · uniform baseline 4.91 bit
| Answer | Count | Share |
|---|---|---|
| 青 | 8 | 30.8% |
| 紫 | 8 | 30.8% |
| 蓝 | 7 | 26.9% |
| 绿 | 3 | 11.5% |
Random letter · English
n = 25 · H = 1.51 bit · uniform baseline 4.70 bit
| Answer | Count | Share |
|---|---|---|
| k | 17 | 68.0% |
| q | 3 | 12.0% |
| g | 2 | 8.0% |
| m | 2 | 8.0% |
| f | 1 | 4.0% |
Random number 1-10 · English
n = 30 · H = 0.00 bit · uniform baseline 3.32 bit
| Answer | Count | Share |
|---|---|---|
| 7 | 30 | 100.0% |
Random number 1-10 · Chinese
n = 27 · H = 0.00 bit · uniform baseline 3.32 bit
| Answer | Count | Share |
|---|---|---|
| 7 | 27 | 100.0% |
Random number 1-100 · English
n = 28 · H = 0.71 bit · uniform baseline 6.64 bit
| Answer | Count | Share |
|---|---|---|
| 42 | 24 | 85.7% |
| 73 | 3 | 10.7% |
| 47 | 1 | 3.6% |
Random number 1-100 · Chinese
n = 24 · H = 0.94 bit · uniform baseline 6.64 bit
| Answer | Count | Share |
|---|---|---|
| 42 | 19 | 79.2% |
| 73 | 3 | 12.5% |
| 47 | 2 | 8.3% |
Want to verify your API really serves GLM 5.1?
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.