# What is token pricing?

> Token pricing is how AI APIs charge: a price per million tokens in and out. How to read a price list, with real October 2026 examples.

Source: https://learn.ynteractive.com/content/glossary/token-pricing · Updated 2026-10-08 · Free from Stepthrough (https://learn.ynteractive.com)

[AI basics](https://learn.ynteractive.com/content/glossary#topic-ai-basics) · [AI glossary](https://learn.ynteractive.com/content/glossary)

Quick answer

Token pricing is the way AI companies charge for API use: a set price per million tokens of text you send in, and a separate, usually higher price per million tokens the model writes back. A token is roughly three quarters of an English word.

Last updated October 8, 2026

Updated October 8, 2026 · By [Robert Breen](https://learn.ynteractive.com/content/about)

## Why it matters for a small business

Token pricing only applies when software calls a model through an API, for example an n8n workflow or an app your developer built. A ChatGPT, Claude or Gemini subscription is a flat monthly price with usage limits instead. Price lists show two numbers per model, input and output, because writing costs the provider more than reading. Bigger models cost more per token, and the gap between a vendor's top tier and its small tier can be 50 to 100 times.

For a small business, the number that matters is cost per job. A short email classification might use a few hundred tokens, so even thousands of them on a small model can cost cents. A long contract sent to a top model, many times a day, is a different budget. Most vendors also offer discounts for batch jobs and repeated prompts. The [API usage billing](https://learn.ynteractive.com/content/glossary/api-usage-billing) entry explains the billing side; this one is about reading the price list.

## Published API prices per million tokens (input / output), as of October 2026

| Model | Input | Output |
| --- | --- | --- |
| Claude Opus 5.5 (Anthropic) | $4 | $20 |
| Claude Haiku 5.5 (Anthropic, prompts up to 100K tokens) | $0.10 | $0.50 |
| GPT-6.1 Sol (OpenAI) | $2 | $10 |
| Gemini 3.1 Pro Preview (Google, prompts up to 200K tokens) | $2 | $12 |

From each vendor's own pricing page. Prices change often and vary by prompt length and discounts, so check the current page before you budget.

## In a real lesson: n8n AI Agent Tutorial: Save Social Media Ideas to Google Sheets

In the [AI Social Media Idea Agent lesson](https://learn.ynteractive.com/content/n8n-ai-agent-tutorial-google-sheets), you add credit at **platform.openai.com** before making a key. The lesson's advice is "Ten dollars goes a long way: you pay per request, not per month," and the narration adds that generating social media ideas costs fractions of a cent per request. Then you pick **gpt-5-mini** in the **Model** dropdown, one of the cheaper models per token.

[Try this lesson free](https://learn.ynteractive.com/modules/n8n-ai-agent) or [read the step-by-step guide](https://learn.ynteractive.com/content/n8n-ai-agent-tutorial-google-sheets).

## Common confusions

### Token pricing vs a subscription

A subscription is a flat monthly fee for a chat app. Token pricing is pay-as-you-go for software. Paying for one never covers the other.

### Input vs output tokens

Input is everything you send, including the system prompt and pasted documents. Output is what the model writes. Long pasted documents raise input; long answers raise output.

## Tips

- Set a monthly spending limit in the vendor's billing settings before you go live.
- Ask for short, structured answers to keep output tokens down.
- Run a test batch and read the usage page before you estimate a monthly cost.

## Related terms

[Token](https://learn.ynteractive.com/content/glossary/token) · [API usage billing](https://learn.ynteractive.com/content/glossary/api-usage-billing) · [OpenAI API](https://learn.ynteractive.com/content/glossary/openai-api) · [Context window](https://learn.ynteractive.com/content/glossary/context-window) · [Model family](https://learn.ynteractive.com/content/glossary/model-family) · [Model router](https://learn.ynteractive.com/content/glossary/model-router)

## More AI basics terms

[Training data](https://learn.ynteractive.com/content/glossary/training-data) · [Vector database](https://learn.ynteractive.com/content/glossary/vector-database) · [AI bias](https://learn.ynteractive.com/content/glossary/ai-bias) · [AI evals](https://learn.ynteractive.com/content/glossary/ai-evals)

## Where you use it: free lessons

- [n8n AI Agent Tutorial: Save Social Media Ideas to Google Sheets](https://learn.ynteractive.com/content/n8n-ai-agent-tutorial-google-sheets) (n8n, 12 min)
- [AI Email Responder: Draft Gmail Replies Automatically with n8n](https://learn.ynteractive.com/content/ai-email-responder-gmail-drafts-n8n) (n8n, 12 min)
- [AI Receipt Extractor: Receipts to Google Sheets with n8n](https://learn.ynteractive.com/content/ai-receipt-extractor-n8n-google-sheets) (n8n, 12 min)

## Frequently asked questions

**How much is a million tokens?**

Roughly 750,000 English words, or well over a thousand pages. Most single business requests use a few hundred to a few thousand tokens.

**Why are output tokens more expensive?**

Generating text takes the model more computing work than reading it, so vendors charge more for each token the model writes back to you.

**Is token pricing the same for every vendor?**

The method is similar, but the prices, discounts and rules for long prompts differ. Compare the cost of your actual job on each vendor's current price list.

[All AI glossary terms, A to Z](https://learn.ynteractive.com/content/glossary) · [Free prompt templates](https://learn.ynteractive.com/content/prompts)