Introducing the New Anthropic Token Counting API

Anthropic has released some exciting beta features in the last couple of days that have largely gone under the radar. One of these was the ability to process PDFs with their models, which can now understand both text and visual content within PDF documents. I’ll maybe write up something on that at a later date.

The other exciting beta feature, and the subject of this article, was the introduction of token counting. Crucially, you can count the tokens in user messages, PDFs and images before you send them to Claude. This is excellent news for those who like to monitor their token usage costs closely.

According to the official announcement from Anthropic (link here),

“The token counting endpoint accepts the same structured list of inputs for creating a message, including support for system prompts, tools, images, and PDFs. The response contains the total number of input tokens.”

And supports the following models,

“Claude 3.5 Sonnet
Claude 3.5 Haiku
Claude 3 Haiku
Claude 3 Opus”