Home/Tools/Mistral Tokenizer

Mistral Tokenizer and Token Counter

Paste text, a prompt, or code to count tokens with the pinned Mistral Medium 3.5 tokenizer, then inspect the exact token pieces and IDs in your browser.

Ready
0tokens
0characters
0words
Tokenized text will appear here as you type.

Your text stays in this browser. Nothing you paste is uploaded.

How counting works

This page runs the pinned Mistral Medium 3.5 tokenizer locally in a browser Worker and returns token pieces and IDs.

Only plain text is encoded. Mistral chat templates, roles, system instructions, tools, and model-specific request formatting are not added.

For request-volume and output-token calculations, see the monthly LLM API cost guide. Using OpenAI? Check the current OpenAI API token prices.

To compare the same English, Chinese, JSON, and code samples, open the reproducible tokenizer comparison.

Count boundary

What the Mistral token count includes

The counter sends the exact editor contents to the local tokenizer. Nothing is estimated or split on spaces.

Text, code, and whitespace

Words, punctuation, indentation, spaces, and line breaks are all part of the string being tokenized.

Tokenizer-produced IDs

The Text and Token IDs views come from the pinned Mistral tokenizer asset, so you can inspect boundaries as well as the total.

Plain text only

The result does not add chat templates, roles, tool definitions, images, or other API request formatting.

View tokenizer source
How StackLens verifies data and calculates cost

Mistral Medium 3.5 tokenizer sourceHigh confidenceChecked 2026-07-19

StackLens pins Mistral's tokenizer JSON and runs it locally in a browser Worker. Chat templates and API request wrappers are not included.

FAQ

Mistral Tokenizer questions

How do I count Mistral tokens online?

Paste text into the editor and the page runs the pinned Mistral Medium 3.5 tokenizer locally. The total, token pieces, and token IDs update from the tokenizer result.

Which Mistral tokenizer does this page use?

It uses the pinned tokenizer JSON for Mistral Medium 3.5 and states that version explicitly.

Does the Mistral tokenizer run locally?

Yes. Tokenization runs in a browser Worker, and the entered text is not uploaded by StackLens.

Can I inspect Mistral token IDs?

Yes. The page displays token pieces and IDs produced by the selected tokenizer asset.

Will Mistral API usage match this number exactly?

Not always. This page counts plain text and does not add chat templates, roles, tools, or other API request formatting.