AI Desk · Model Identity · Evidence object
Tokenizer signature tk_9c4e2396
An exact tokenizer signature: every model listed below returned the same marginal prompt-token count on all 50 test strings. Sharing a signature is strong evidence of a shared vocabulary. This signature sits in the Qwen · group A group.
- Members
- 1
- First observed
- 20 August 2026
- Last observed
- 22 August 2026
- Group
- Qwen · group A
Models with this signature
Template variants of the same tokenizer
These signatures differ from this one by a uniform one-token shift on the affected strings: the same vocabulary behind a different chat template.
The measured vector
Marginal prompt-token cost per string, as measured on Xiaomi: MiMo-V2.5-ProAll 50 rows
English and whitespace
en-prose12
en-long14
spaces-201
spaces-601
tabs-201
newlines-202
mixed-ws3
Digits
digits-97
digits-1210
digits-3028
digits-sep10
float-long20
CJK
zh-common8
zh-long10
zh-rare15
ja-kana6
ja-kanji11
ko15
Other scripts
ru14
ar6
he7
hi27
th10
el26
Emoji
emoji-basic3
emoji-skin8
emoji-zwj-family8
emoji-zwj-x328
emoji-flags6
emoji-prof12
Rare Unicode
math24
boxdraw15
combining3
cjk-ext-b10
surrogates20
zalgo34
rtl-mix5
Code
py-code23
py-indent13
json23
html14
regex42
camel3
snake4
Repetition and encodings
rare-word-x529
repeat-tok20
base6434
hex15
url17
uuid34
Method: tokenizer fingerprinting, v1-50.