GPT-4o token counter
Type or paste a prompt to count GPT-4o tokens and estimate what it costs to run — free, live, and private.
Try:
Counts include each model's own chat-template overhead.
128 charsGPT-4o
OpenAI · 128k context…tokens
About GPT-4o tokenization
- Vendor
- OpenAI
- Context window
- 128k tokens
- Pricing (per 1M tokens)
- $2.5 input · $10 output (as of 2026-07-07)
- How we count
- the o200k_base encoding, running entirely in your browser
FAQ
How accurate is this GPT-4o token count?
Exact — it uses the o200k_base encoding, running entirely in your browser, the same tokenizer the model uses in production.
Does my prompt leave the browser?
No. Tokenization runs entirely in your browser.
What does a GPT-4o request cost?
Input tokens × $2.5/1M plus output tokens × $10/1M (standard tier, as of 2026-07-07). The calculator above does this for your actual prompt.