EMAIL → TOKENS

How many tokens is an email?

Last updated

A typical work email of about 120 words is roughly 160 tokens. Short replies run 30-60 tokens, long formal emails 400-700, and a full quoted thread can easily exceed 2,000 tokens.

QUICK CONVERTER

≈ Words

120

≈ Tokens

160

Estimates use ~1.33 tokens per English word. For an exact count of real text — including code, JSON, and other languages — paste it into the token counterand choose your model.

The detail

Email is one of the most common AI workloads — drafting, summarizing, classifying, and extracting from messages. A standalone 120-word email is about 160 tokens, but real emails rarely arrive standalone: signatures, disclaimers, and quoted history often triple the token count of what a human would consider "the email".

For high-volume processing the overhead matters more than the message. Stripping quoted threads and signatures before sending messages to a model routinely cuts token spend by 50-70% with no loss of information, because the model only needs the newest content plus a compact summary of the thread.

At $0.5-$5 per million input tokens, classifying or summarizing a single email costs thousandths of a cent — it is the millions-of-emails scale where tokenization discipline turns into real savings.

Email lengths in tokens

Email~Words~Tokens
Short reply (~30 words)3040
Average email (~120 words)120160
Detailed email (~300 words)300399
Formal letter (~500 words)500665
Email + quoted thread (~1,500 words)1,5001,995

FAQ

How many tokens is an email? — FAQ

How many tokens does an email signature add?

A typical signature block with name, title, company, phone, and legal disclaimer adds 50-150 tokens per message — often more than a short reply itself. Strip signatures before bulk processing.

What does it cost to summarize 10,000 emails with AI?

At ~400 tokens per email including partial thread context and a ~50-token summary each, about 4.5M total tokens. On a budget model at $0.3/M input and $1.5/M output, roughly $2-3 for the whole batch.

Should I send the whole email thread to the model?

Usually no. Send the newest message plus a running summary of the thread. Quoted history grows quadratically over a long exchange and mostly repeats what the model has already processed.