AI Desk · Model Identity · Evidence object
Tokenizer signature tk_5410b38e
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
- 27
- First observed
- 20 August 2026
- Last observed
- 21 August 2026
- Group
- Qwen · group A
Models with this signature
ByteDance: UI-TARS 7B bytedance/ui-tars-1.5-7bBytedanceDots Studio: Dots3-Note Preview (free)dots-studio/dots-3-note-preview:freeDots StudioQwen2.5 72B Instructqwen/qwen-2.5-72b-instructAlibaba (Qwen)Qwen: Qwen2.5 7B Instructqwen/qwen-2.5-7b-instructAlibaba (Qwen)Qwen: Qwen-Plusqwen/qwen-plusAlibaba (Qwen)Qwen: Qwen Plus 0728qwen/qwen-plus-2025-07-28Alibaba (Qwen)Qwen: Qwen3 14Bqwen/qwen3-14bAlibaba (Qwen)Qwen: Qwen3 235B A22B Instruct 2507qwen/qwen3-235b-a22b-2507Alibaba (Qwen)Qwen: Qwen3 235B A22B Thinking 2507qwen/qwen3-235b-a22b-thinking-2507Alibaba (Qwen)Qwen: Qwen3 30B A3Bqwen/qwen3-30b-a3bAlibaba (Qwen)Qwen: Qwen3 30B A3B Instruct 2507qwen/qwen3-30b-a3b-instruct-2507Alibaba (Qwen)Qwen: Qwen3 30B A3B Thinking 2507qwen/qwen3-30b-a3b-thinking-2507Alibaba (Qwen)Qwen: Qwen3 32Bqwen/qwen3-32bAlibaba (Qwen)Qwen: Qwen3 8Bqwen/qwen3-8bAlibaba (Qwen)Qwen: Qwen3 Coder 480B A35Bqwen/qwen3-coderAlibaba (Qwen)Qwen: Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instructAlibaba (Qwen)Qwen: Qwen3 Coder Flashqwen/qwen3-coder-flashAlibaba (Qwen)Qwen: Qwen3 Coder Nextqwen/qwen3-coder-nextAlibaba (Qwen)Qwen: Qwen3 Next 80B A3B Instructqwen/qwen3-next-80b-a3b-instructAlibaba (Qwen)Qwen: Qwen3 Next 80B A3B Thinkingqwen/qwen3-next-80b-a3b-thinkingAlibaba (Qwen)Qwen: Qwen3 VL 235B A22B Instructqwen/qwen3-vl-235b-a22b-instructAlibaba (Qwen)Qwen: Qwen3 VL 30B A3B Instructqwen/qwen3-vl-30b-a3b-instructAlibaba (Qwen)Qwen: Qwen3 VL 30B A3B Thinkingqwen/qwen3-vl-30b-a3b-thinkingAlibaba (Qwen)Qwen: Qwen3 VL 32B Instructqwen/qwen3-vl-32b-instructAlibaba (Qwen)Qwen: Qwen3 VL 8B Instructqwen/qwen3-vl-8b-instructAlibaba (Qwen)Qwen: Qwen3 VL 8B Thinkingqwen/qwen3-vl-8b-thinkingAlibaba (Qwen)Xiaomi: MiMo-V2.5xiaomi/mimo-v2.5Xiaomi
The measured vector
Marginal prompt-token cost per string, as measured on ByteDance: UI-TARS 7BAll 50 rows
English and whitespace
en-prose14
en-long16
spaces-203
spaces-603
tabs-203
newlines-204
mixed-ws5
Digits
digits-99
digits-1212
digits-3030
digits-sep12
float-long22
CJK
zh-common10
zh-long12
zh-rare17
ja-kana8
ja-kanji13
ko17
Other scripts
ru16
ar8
he9
hi29
th12
el28
Emoji
emoji-basic5
emoji-skin10
emoji-zwj-family10
emoji-zwj-x330
emoji-flags8
emoji-prof14
Rare Unicode
math26
boxdraw17
combining5
cjk-ext-b12
surrogates22
zalgo36
rtl-mix7
Code
py-code25
py-indent15
json25
html16
regex44
camel5
snake6
Repetition and encodings
rare-word-x531
repeat-tok22
base6436
hex17
url19
uuid36
Method: tokenizer fingerprinting, v1-50. Raw data: fingerprints.json · index.json · test-strings.json.