# AI token cost calculator: Claude vs OpenAI prices

> Enter tokens in and out and calls per month, and compare the cost on every current Claude and OpenAI model, with prompt caching. Paste text to estimate its tokens. Official prices, free, in your browser.

Canonical: https://dejo.run/tools/ai-token-cost-calculator

> For AI agents: the index of this site is https://dejo.run/llms.txt, and every page is also served as markdown (add .md to its address, or send Accept: text/markdown).

**The short answer:** enter how many tokens go in and come out of each call and how many calls you make a month, and see the cost on every current Claude and OpenAI model side by side. Paste text to estimate its tokens first.


It runs in your browser; nothing you paste is sent anywhere. Prices are the vendors' standard API prices, checked on 10 November 2026.

## Prices per million tokens

| Model | Input | Cached input | Output |
|---|---|---|---|
| Claude Fable 5.2 | $8 | $0.20 | $40 |
| Claude Opus 5.5 | $4 | $0.20 | $20 |
| Claude Sonnet 5.5 | $2 | $0.10 | $10 |
| Claude Haiku 5.5 (prompts up to 100K) | $0.10 | $0.01 | $0.50 |
| gpt-6-astra | $10 | $1 | $50 |
| gpt-6.1-sol | $2 | $0.10 | $10 |
| gpt-6.1-terra | $1 | $0.05 | $5 |
| gpt-6-luna | $0.10 | $0.01 | $0.50 |
| gpt-5.4-mini | $0.75 | $0.075 | $4.50 |
| gpt-5-nano | $0.05 | $0.005 | $0.40 |

Output costs five to eight times as much as input on every model here, so a long answer costs more than a prompt of the same length.

## The calls you do not need to make

Agents that use websites pay for every page they read, on every run. A task done every week reads the same pages every week. [Dejo](https://dejo.run/) learns a website task once and replays it later with zero model calls, so the repeat runs skip most of that cost; see the [Claude Code cost calculator](https://dejo.run/tools/claude-code-cost-calculator) for what that comes to.

## Sources

- Claude prices: [claude.com/pricing](https://www.claude.com/pricing)
- OpenAI prices: [developers.openai.com/api/docs/pricing](https://developers.openai.com/api/docs/pricing)
- Tokens per character: [Anthropic glossary](https://platform.claude.com/docs/en/about-claude/glossary)

## FAQ

## FAQ

### How many tokens is my text?

For current Claude models, Anthropic's own rule is that earlier models used about one token per 3.5 English characters and current models make about 30% more tokens for the same text. Paste your text to get that estimate. For an exact count, use the vendor's token counting API with the model you plan to use.

### Why does cached input cost less?

When the start of a prompt repeats across calls (instructions, a long document, tool definitions), both Anthropic and OpenAI can read it from a prompt cache at a fraction of the normal input price. The calculator lets you set what share of the input comes from the cache.

### Which model is cheapest?

Per token, the smallest models: Claude Haiku 5.5 and gpt-5-nano or gpt-6-luna. The cheapest model for a job is the smallest one that still does it well, and the cheapest call is the one you do not need to make.

## Related

- [Dejo: browser automation MCP that remembers websites](https://dejo.run/)
- [Dejo MCP server setup: install in a minute](https://dejo.run/docs)
