bytedance-seed/seed-1.6-flash behavioral LLM fingerprint
This page breaks down the exact reference that VerifyLLMAPI uses for bytedance-seed/seed-1.6-flash: 116 normalized answers across eight low-entropy probes, plus its nearest enrolled models by mean Jensen–Shannon divergence.
Fingerprint summary
- Reference answers
- 116
- Samples per cell
- 13–15
- Nearest model
- qwen/qwen3-next-80b-a3b-instruct
- Nearest mean JSD
- 0.707884
The most concentrated cell was animal-random|zh: 章鱼 held 78.6% of its answers and entropy measured 1.089 bits. The broadest cell was animal-random|ar at 3.457 bits.
Against the closest enrolled model, animal-random|en contributed the largest cell distance at JSD 1.000000. 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 116/120 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 · 13/15 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 · 14/15 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 · 15/15 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 · 14/15 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. 13 valid reference answers produced 6 normalized values. The mode was axolotl at 4/13 (30.8%); entropy was 2.412 bits.
| Normalized response | Count | Share |
|---|---|---|
axolotl |
4 | 30.8% |
fennec |
3 | 23.1% |
koala |
2 | 15.4% |
quokka |
2 | 15.4% |
anteater |
1 | 7.7% |
fossa |
1 | 7.7% |
2. word-random|en
Random word, English. 14 valid reference answers produced 6 normalized values. The mode was apple at 7/14 (50.0%); entropy was 2.118 bits.
| Normalized response | Count | Share |
|---|---|---|
apple |
7 | 50.0% |
banana |
2 | 14.3% |
table |
2 | 14.3% |
book |
1 | 7.1% |
elephant |
1 | 7.1% |
giraffe |
1 | 7.1% |
3. animal-random|ru
Random animal, Russian. 15 valid reference answers produced 8 normalized values. The mode was собака at 8/15 (53.3%); entropy was 2.307 bits.
| Normalized response | Count | Share |
|---|---|---|
собака |
8 | 53.3% |
завтра |
1 | 6.7% |
зебра |
1 | 6.7% |
лиса |
1 | 6.7% |
словарь |
1 | 6.7% |
слон |
1 | 6.7% |
с龙马 |
1 | 6.7% |
سобака |
1 | 6.7% |
4. animal-random|ar
Random animal, Arabic. 15 valid reference answers produced 12 normalized values. The mode was أرنب at 3/15 (20.0%); entropy was 3.457 bits.
| Normalized response | Count | Share |
|---|---|---|
أرنب |
3 | 20.0% |
أسد |
2 | 13.3% |
أضحك |
1 | 6.7% |
أفريقي |
1 | 6.7% |
المرقوق |
1 | 6.7% |
الملح |
1 | 6.7% |
النمر |
1 | 6.7% |
ال仿若 |
1 | 6.7% |
دجاج |
1 | 6.7% |
قرفدش |
1 | 6.7% |
كلب |
1 | 6.7% |
نمر |
1 | 6.7% |
5. letter-random|zh
Random character, Chinese. 15 valid reference answers produced 6 normalized values. The mode was 一 at 9/15 (60.0%); entropy was 1.872 bits.
| Normalized response | Count | Share |
|---|---|---|
一 |
9 | 60.0% |
山 |
2 | 13.3% |
人 |
1 | 6.7% |
日 |
1 | 6.7% |
林 |
1 | 6.7% |
水 |
1 | 6.7% |
6. animal-random|zh
Random animal, Chinese. 14 valid reference answers produced 4 normalized values. The mode was 章鱼 at 11/14 (78.6%); entropy was 1.089 bits.
| Normalized response | Count | Share |
|---|---|---|
章鱼 |
11 | 78.6% |
树懒 |
1 | 7.1% |
树袋熊 |
1 | 7.1% |
鸵鸟 |
1 | 7.1% |
7. letter-random|ar
Random letter, Arabic. 15 valid reference answers produced 8 normalized values. The mode was ن at 6/15 (40.0%); entropy was 2.556 bits.
| Normalized response | Count | Share |
|---|---|---|
ن |
6 | 40.0% |
ك |
3 | 20.0% |
ء |
1 | 6.7% |
ب |
1 | 6.7% |
ر |
1 | 6.7% |
ق |
1 | 6.7% |
م |
1 | 6.7% |
ي |
1 | 6.7% |
8. color-random|zh
Random color, Chinese. 15 valid reference answers produced 6 normalized values. The mode was 紫色 at 7/15 (46.7%); entropy was 2.146 bits.
| Normalized response | Count | Share |
|---|---|---|
紫色 |
7 | 46.7% |
天蓝色 |
3 | 20.0% |
靛蓝 |
2 | 13.3% |
橙色 |
1 | 6.7% |
蓝色 |
1 | 6.7% |
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 | qwen/qwen3-next-80b-a3b-instruct |
0.707884 |
| 2 | nvidia/nemotron-3-super-120b-a12b |
0.743354 |
| 3 | openai/gpt-3.5-turbo |
0.758953 |
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: 334a306d3a6c44da09bfed2fc92cc87212a50299a4899cc80e25db8e3eb1d676. This hash covers the model ID, eight cell metrics, nearest references, and strongest cell difference in the generated atlas.