bytedance-seed/seed-2.0-mini behavioral LLM fingerprint
This page breaks down the exact reference that VerifyLLMAPI uses for bytedance-seed/seed-2.0-mini: 119 normalized answers across eight low-entropy probes, plus its nearest enrolled models by mean Jensen–Shannon divergence.
Fingerprint summary
- Reference answers
- 119
- Samples per cell
- 14–15
- Nearest model
- moonshotai/kimi-k2.6
- Nearest mean JSD
- 0.706843
The most concentrated cell was animal-random|en: platypus held 73.3% of its answers and entropy measured 1.237 bits. The broadest cell was animal-random|ru at 3.379 bits.
Against the closest enrolled model, letter-random|zh contributed the largest cell distance at JSD 0.933333. This is a descriptive comparison, not proof that either API route serves a named checkpoint.
Completeness: the author dataset allows up to 15 valid answers per selected cell in reference half A. This model contains 119/120 across the eight cells. We show the actual denominator for every share.
The eight test requests
The research pipeline sent each user prompt with a one-answer system constraint, temperature 1, a 16-token output cap, and reasoning disabled. VerifyLLMAPI uses the same prompt text and normalization target.
| Cell | Task | Exact request text | Normalization and n |
|---|---|---|---|
animal-random|en |
Random animal · English | Name a random animal.System constraintAnswer with a single word or a single number only. Output nothing else: no explanation, no punctuation, no quotation marks. |
word · 15/15 reference samples |
word-random|en |
Random word · English | Name a random word.System constraintAnswer with a single word or a single number only. Output nothing else: no explanation, no punctuation, no quotation marks. |
word · 15/15 reference samples |
animal-random|ru |
Random animal · Russian | Назови случайное животное.System constraintОтвечай только одним словом или одним числом. Не выводи ничего больше: без объяснений, без знаков препинания, без кавычек. |
word · 14/15 reference samples |
animal-random|ar |
Random animal · Arabic | اذكر حيواناً عشوائياً.System constraintأجب بكلمة واحدة أو رقم واحد فقط. لا تكتب أي شيء آخر: لا شرح ولا علامات ترقيم ولا علامات اقتباس. |
word · 15/15 reference samples |
letter-random|zh |
Random character · Chinese | 说出一个随机的汉字。System constraint只用一个词或一个数字回答。不要输出任何其他内容：不要解释，不要标点符号，不要引号。 |
grapheme · 15/15 reference samples |
animal-random|zh |
Random animal · Chinese | 说出一个随机的动物。System constraint只用一个词或一个数字回答。不要输出任何其他内容：不要解释，不要标点符号，不要引号。 |
word · 15/15 reference samples |
letter-random|ar |
Random letter · Arabic | اذكر حرفاً عشوائياً من الحروف الأبجدية.System constraintأجب بكلمة واحدة أو رقم واحد فقط. لا تكتب أي شيء آخر: لا شرح ولا علامات ترقيم ولا علامات اقتباس. |
grapheme · 15/15 reference samples |
color-random|zh |
Random color · Chinese | 说出一个随机的颜色。System constraint只用一个词或一个数字回答。不要输出任何其他内容：不要解释，不要标点符号，不要引号。 |
word · 15/15 reference samples |
Response distributions
These are normalized reference-half counts from the published OpenRouter dataset. The tables contain aggregate values, not user data or a live VerifyLLMAPI scan.
1. animal-random|en
Random animal, English. 15 valid reference answers produced 4 normalized values. The mode was platypus at 11/15 (73.3%); entropy was 1.237 bits.
| Normalized response | Count | Share |
|---|---|---|
platypus |
11 | 73.3% |
axolotl |
2 | 13.3% |
kaka |
1 | 6.7% |
quokka |
1 | 6.7% |
2. word-random|en
Random word, English. 15 valid reference answers produced 7 normalized values. The mode was sunflower at 6/15 (40.0%); entropy was 2.423 bits.
| Normalized response | Count | Share |
|---|---|---|
sunflower |
6 | 40.0% |
umbrella |
3 | 20.0% |
whimsical |
2 | 13.3% |
cinnamon |
1 | 6.7% |
octopus |
1 | 6.7% |
spoon |
1 | 6.7% |
umbrage |
1 | 6.7% |
3. animal-random|ru
Random animal, Russian. 14 valid reference answers produced 11 normalized values. The mode was коала at 2/14 (14.3%); entropy was 3.379 bits.
| Normalized response | Count | Share |
|---|---|---|
коала |
2 | 14.3% |
кошка |
2 | 14.3% |
крокодил |
2 | 14.3% |
американский |
1 | 7.1% |
калан |
1 | 7.1% |
котик |
1 | 7.1% |
курица |
1 | 7.1% |
ламантин |
1 | 7.1% |
лемур |
1 | 7.1% |
носорог |
1 | 7.1% |
пингвин |
1 | 7.1% |
4. animal-random|ar
Random animal, Arabic. 15 valid reference answers produced 8 normalized values. The mode was أسد at 6/15 (40.0%); entropy was 2.606 bits.
| Normalized response | Count | Share |
|---|---|---|
أسد |
6 | 40.0% |
الأسد |
2 | 13.3% |
قطة |
2 | 13.3% |
الذئب |
1 | 6.7% |
حمير |
1 | 6.7% |
دب |
1 | 6.7% |
دلفين |
1 | 6.7% |
فيل |
1 | 6.7% |
5. letter-random|zh
Random character, Chinese. 15 valid reference answers produced 12 normalized values. The mode was 明 at 4/15 (26.7%); entropy was 3.374 bits.
| Normalized response | Count | Share |
|---|---|---|
明 |
4 | 26.7% |
亮 |
1 | 6.7% |
喃 |
1 | 6.7% |
川 |
1 | 6.7% |
晴 |
1 | 6.7% |
暗 |
1 | 6.7% |
林 |
1 | 6.7% |
森 |
1 | 6.7% |
的 |
1 | 6.7% |
苍 |
1 | 6.7% |
闲 |
1 | 6.7% |
默 |
1 | 6.7% |
6. animal-random|zh
Random animal, Chinese. 15 valid reference answers produced 8 normalized values. The mode was 海狸 at 3/15 (20.0%); entropy was 2.823 bits.
| Normalized response | Count | Share |
|---|---|---|
海狸 |
3 | 20.0% |
考拉 |
3 | 20.0% |
鸭嘴兽 |
3 | 20.0% |
企鹅 |
2 | 13.3% |
海獭 |
1 | 6.7% |
犰狳 |
1 | 6.7% |
章鱼 |
1 | 6.7% |
针鼹 |
1 | 6.7% |
7. letter-random|ar
Random letter, Arabic. 15 valid reference answers produced 7 normalized values. The mode was ب at 5/15 (33.3%); entropy was 2.499 bits.
| Normalized response | Count | Share |
|---|---|---|
ب |
5 | 33.3% |
ز |
3 | 20.0% |
م |
3 | 20.0% |
س |
1 | 6.7% |
ط |
1 | 6.7% |
غ |
1 | 6.7% |
ف |
1 | 6.7% |
8. color-random|zh
Random color, Chinese. 15 valid reference answers produced 7 normalized values. The mode was 天蓝色 at 6/15 (40.0%); entropy was 2.340 bits.
| Normalized response | Count | Share |
|---|---|---|
天蓝色 |
6 | 40.0% |
靛蓝 |
4 | 26.7% |
天蓝 |
1 | 6.7% |
天青色 |
1 | 6.7% |
蔚蓝 |
1 | 6.7% |
藏青 |
1 | 6.7% |
藏青色 |
1 | 6.7% |
Three nearest enrolled references
We compute base-2 JSD for each of the eight categorical distributions, then average the eight distances. A lower value means more similar answer frequencies within this product reference. It does not identify an unknown route by itself.
| Rank | Reference model | Mean 8-cell JSD |
|---|---|---|
| 1 | moonshotai/kimi-k2.6 |
0.706843 |
| 2 | moonshotai/kimi-k2.5 |
0.711681 |
| 3 | moonshotai/kimi-k2-0905 |
0.743293 |
How many live requests does verification use?
| Profile | Cells × samples | Fresh model requests | Same-data EER |
|---|---|---|---|
| Quick, default complete check | 4 × 5 | 20 | 11.8% |
| Standard | 8 × 5 | 40 | 8.1% |
| Paper budget | 8 × 15 | 120 | 6.7% |
The current-Agent workflow starts with quick. Each answer comes from a fresh Codex or Claude Code process that reuses host-managed login. These requests can consume host allowance or provider balance.
What this fingerprint can and cannot show
- The reference records behavior observed through OpenRouter in the author dataset. Provider wrappers, model updates, decoding, and date can move a distribution.
- The author collection set temperature 1 and disabled reasoning. Current-Agent mode labels sampling as a host default because supported Agent CLIs expose no temperature flag.
- The eight high-gap cells and product thresholds use the same public cohort. Their EER values are calibration results, not an independent production accuracy claim.
- A nearest reference is a similarity result. VerifyLLMAPI returns INCONCLUSIVE when the claimed route lacks an enrolled fingerprint.
Primary sources, license, and reproducibility
- Tomas Bruckner, One Token Is Enough: Fingerprinting and Verifying Large Language Models from Single-Token Output Distributions, arXiv:2607.10252v1.
- Author dataset and derived results, DOI 10.5281/zenodo.21278557, CC BY 4.0.
- Author collection and analysis software, DOI 10.5281/zenodo.21278793, MIT.
- VerifyLLMAPI atlas JSON, built from reference
pamela-openrouter-2026-07-8cell-ref-a, SHA-25651f3f63cbcfb56b151055d2d10e36a9cc2f0644cc68b580784853d68951fa866.
Page evidence SHA-256: fd77df82d84f8d9ccb71d02b3f1acbbd4b3231f04cbb945315ea49b00d52cafd. This hash covers the model ID, eight cell metrics, nearest references, and strongest cell difference in the generated atlas.