Gemini token counter
Paste a prompt or document to estimate its token count and cost on Google's Gemini models. Gemini uses its own tokenizer, so counts are estimates based on the same text.
How to use the Gemini token counter
- Paste your prompt or document.
- Read the estimate in the Gemini rows of the table.
- Set the expected answer length and the number of requests.
- Compare the Gemini models' cost and context use.
What you get
- Token estimates for current Gemini models
- Promotional prices noted with their end date
- Cost per request and for many requests
- Context window check
About the Gemini token counter
Estimates, clearly marked
Google's tokenizer is not available for offline counting, so the count comes from the same text measured with OpenAI's tokenizer. For exact numbers, use the countTokens method in the Gemini API.
Large documents
Gemini models accept very long inputs. Paste a whole report or transcript to see how much of the window it uses before you send it.
Frequently asked questions
- Why does Gemini show two prices?
- Some Gemini prices are promotional for a period. The note under the model name shows the regular price that follows.
- Can I count images or audio?
- No, this page counts text. Images, audio and video are billed in other ways.
- Is my text uploaded?
- No. Everything runs in your browser.