AI Desk · AI Fingerprinting · Compare
Mistral: Mistral Medium 3.5 · Mistral: Voxtral Small 24B 2507
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.
Mistral: Mistral Medium 3.5 recordMistral: Voxtral Small 24B 2507 recordOpen in the interactive bench
Tokenizer fingerprint
50 of 50 mutually clean strings agree| Test string | Mistral: Mistral Medium 3.5 | Mistral: Voxtral Small 24B 2507 | Verdict |
|---|---|---|---|
| en-prose | 14 | 14 | match |
| en-long | 18 | 18 | match |
| spaces-20 | 3 | 3 | match |
| spaces-60 | 3 | 3 | match |
| tabs-20 | 4 | 4 | match |
| newlines-20 | 7 | 7 | 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 | 16 | 16 | match |
| zh-long | 20 | 20 | match |
| zh-rare | 20 | 20 | match |
| ja-kana | 11 | 11 | match |
| ja-kanji | 13 | 13 | match |
| ko | 13 | 13 | match |
| ru | 12 | 12 | match |
| ar | 9 | 9 | match |
| he | 10 | 10 | match |
| hi | 14 | 14 | match |
| th | 17 | 17 | match |
| el | 13 | 13 | match |
| emoji-basic | 20 | 20 | match |
| emoji-skin | 40 | 40 | match |
| emoji-zwj-family | 19 | 19 | match |
| emoji-zwj-x3 | 57 | 57 | match |
| emoji-flags | 32 | 32 | match |
| emoji-prof | 35 | 35 | match |
| math | 33 | 33 | match |
| boxdraw | 30 | 30 | match |
| combining | 14 | 14 | match |
| cjk-ext-b | 16 | 16 | match |
| surrogates | 53 | 53 | match |
| zalgo | 36 | 36 | match |
| rtl-mix | 6 | 6 | match |
| py-code | 25 | 25 | match |
| py-indent | 15 | 15 | match |
| json | 25 | 25 | match |
| html | 16 | 16 | match |
| regex | 49 | 49 | match |
| camel | 6 | 6 | match |
| snake | 7 | 7 | match |
| rare-word-x5 | 36 | 36 | match |
| repeat-tok | 41 | 41 | match |
| base64 | 39 | 39 | match |
| hex | 19 | 19 | match |
| url | 19 | 19 | 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 | Mistral: Mistral Medium 3.5 | Mistral: Voxtral Small 24B 2507 | Verdict |
|---|---|---|---|
| Context length | 262144 | 32768 | × differs |
| Max output tokens | 209715 | 26214 | × differs |
| Supported parameters | frequency_penalty, include_reasoning, max_tokens, presence_penalty, reasoning, reasoning_effort, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_p | frequency_penalty, max_tokens, presence_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_p | × differs |
| Defaults | {"temperature":null,"top_p":null,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null} | {"temperature":0.2,"top_p":0.95,"frequency_penalty":null} | × differs |
| Reasoning contract | {"mandatory":false,"supported_efforts":["high","none"],"default_effort":"high"} | – | × differs |
| Modalities | text+image+file->text | text+file+audio->text | × differs |
Method: tokenizer fingerprinting, v1, 50 strings. This page is static and dated; the interactive bench compares any two of the 338 measured models.