AI Desk · Model Identity · Evidence object
Tokenizer signature tk_d3c6ae80
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 Aion Labs · group A group.
- Members
- 2
- First observed
- 3 September 2026
- Last observed
- 3 September 2026
- Group
- Aion Labs · group A
Models with this signature
The measured vector
Marginal prompt-token cost per string, as measured on AionLabs: Aion-2.0All 50 rows
English and whitespace
en-prose10
en-long12
spaces-201
spaces-601
tabs-201
newlines-201
mixed-ws3
Digits
digits-91
digits-121
digits-306
digits-sep3
float-long5
CJK
zh-common5
zh-long6
zh-rare14
ja-kana5
ja-kanji7
ko9
Other scripts
ru9
ar6
he10
hi14
th7
el12
Emoji
emoji-basic6
emoji-skin10
emoji-zwj-family7
emoji-zwj-x329
emoji-flags12
emoji-prof15
Rare Unicode
math18
boxdraw23
combining6
cjk-ext-b12
surrogates29
zalgo32
rtl-mix2
Code
py-code21
py-indent11
json16
html12
regex43
camel3
snake4
Repetition and encodings
rare-word-x522
repeat-tok37
base6430
hex8
url13
uuid23
Method: tokenizer fingerprinting, v1-50.