Question 1

Which tokenizer does Cohere use?

Accepted Answer

Cohere uses its own tokenizer, which is not publicly available to run in a browser. This counter estimates counts with OpenAI's cl100k_base encoding, which is typically within roughly 10 to 20 percent of the real count for English text.

Question 2

Are these Cohere token counts exact or estimated?

Accepted Answer

Estimated. Cohere has not released its tokenizer for browser use, so this tool approximates counts with OpenAI's cl100k_base encoding. Estimates are usually close for English text but can drift for code, JSON, and non-English languages. Estimated counts are marked with a tilde (~).

Question 3

How is Cohere API cost calculated?

Accepted Answer

Cost equals tokens divided by one million, multiplied by the model's per-million-token rate. Each Cohere model has separate input and output rates, and this page shows both: what your text would cost as a prompt (input) and what the same number of tokens would cost as a response (output).

Question 4

Why do token counts matter for Cohere API costs?

Accepted Answer

Cohere bills API usage per token, not per word or character. A token is roughly 4 characters or about three quarters of a word in English. Knowing the token count of your prompts lets you predict spend, stay inside context windows, and compare Cohere models like Command A+ and Command A before you commit.

Model	Tokens	Input Cost	Output Cost
Command A+	—	—	—
Command A	—	—	—

Model	Input /1M	Output /1M	Context
Command A+	$2.50	$10.00	128K
Command A	$2.50	$10.00	128K

Cohere Token Counter

Cohere Model Pricing

Frequently Asked Questions

Related Tools