AI Desk · AI Fingerprinting · Compare
ByteDance Seed: Seed 1.6 · ByteDance Seed: Seed-2.0-Lite
Identity similarity: highExact tokenizer agreement · 50/50Exact agreement across every measured string is one strong signal of shared tokenizer lineage. The published settings differ, so the second signal the methodology needs for very high is absent here; the table below shows where they part.
ByteDance Seed: Seed 1.6 recordByteDance Seed: Seed-2.0-Lite recordOpen in the interactive bench
Tokenizer fingerprint
50 of 50 mutually clean strings agree| Test string | ByteDance Seed: Seed 1.6 | ByteDance Seed: Seed-2.0-Lite | Verdict |
|---|---|---|---|
| en-prose | 14 | 14 | match |
| en-long | 16 | 16 | match |
| spaces-20 | 3 | 3 | match |
| spaces-60 | 3 | 3 | match |
| tabs-20 | 4 | 4 | match |
| newlines-20 | 4 | 4 | match |
| mixed-ws | 7 | 7 | match |
| digits-9 | 9 | 9 | match |
| digits-12 | 12 | 12 | match |
| digits-30 | 30 | 30 | match |
| digits-sep | 12 | 12 | match |
| float-long | 22 | 22 | match |
| zh-common | 11 | 11 | match |
| zh-long | 13 | 13 | match |
| zh-rare | 18 | 18 | match |
| ja-kana | 18 | 18 | match |
| ja-kanji | 13 | 13 | match |
| ko | 13 | 13 | match |
| ru | 13 | 13 | match |
| ar | 12 | 12 | match |
| he | 13 | 13 | match |
| hi | 17 | 17 | match |
| th | 16 | 16 | match |
| el | 14 | 14 | match |
| emoji-basic | 9 | 9 | match |
| emoji-skin | 14 | 14 | match |
| emoji-zwj-family | 11 | 11 | match |
| emoji-zwj-x3 | 33 | 33 | match |
| emoji-flags | 16 | 16 | match |
| emoji-prof | 19 | 19 | match |
| math | 30 | 30 | match |
| boxdraw | 29 | 29 | match |
| combining | 5 | 5 | match |
| cjk-ext-b | 16 | 16 | match |
| surrogates | 41 | 41 | match |
| zalgo | 36 | 36 | match |
| rtl-mix | 6 | 6 | match |
| py-code | 26 | 26 | match |
| py-indent | 18 | 18 | match |
| json | 25 | 25 | match |
| html | 16 | 16 | match |
| regex | 50 | 50 | match |
| camel | 7 | 7 | match |
| snake | 8 | 8 | match |
| rare-word-x5 | 31 | 31 | match |
| repeat-tok | 41 | 41 | match |
| base64 | 33 | 33 | match |
| hex | 19 | 19 | match |
| url | 20 | 20 | match |
| uuid | 36 | 36 | match |
Values are the marginal prompt-token cost of each string with the model's fixed overhead subtracted. An x marks a row excluded as corrupt for that model.
API surface
The declared serving contracts differ| Field | ByteDance Seed: Seed 1.6 | ByteDance Seed: Seed-2.0-Lite | Verdict |
|---|---|---|---|
| Context length | 262144 | 262144 | ✓ match |
| Max output tokens | 32768 | 131072 | × differs |
| Supported parameters | frequency_penalty, include_reasoning, max_tokens, reasoning, response_format, stop, structured_outputs, temperature, tool_choice, tools, top_p | frequency_penalty, include_reasoning, max_tokens, reasoning, reasoning_effort, response_format, stop, structured_outputs, temperature, tool_choice, tools, top_p | × differs |
| Defaults | {"temperature":null,"top_p":null,"frequency_penalty":null} | {"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null} | × differs |
| Reasoning contract | {"mandatory":false} | {"mandatory":false,"supported_efforts":["high","medium","low","minimal"],"default_effort":"medium"} | × differs |
| Modalities | text+image+video->text | text+image+video->text | ✓ match |
Method: tokenizer fingerprinting, v1, 50 strings. This page is static and dated; the interactive bench compares any two of the 338 measured models.