OpenAI Token Cost Calculator
See exactly what GPT-4o, o1, and other OpenAI models will cost for your app — before you build.
Select your model
GPT-4o
Best for most tasks
GPT-4o mini
Fastest & cheapest OpenAI option
o1
Advanced reasoning
o1-mini
Fast reasoning model
GPT-3.5 Turbo
Legacy fast model
Set your usage
How many tokens are in your prompt or message
How many tokens the AI writes back
Review results
Cost as you grow
| Users | Requests / Mo | Monthly | Yearly |
|---|---|---|---|
| 100 | 3.0M | $9,750.00 | $117.0K |
| 1.0K | 30.0M | $97,500.00 | $1.2M |
| 10.0K | 300.0M | $975.0K | $11.7M |
| 100.0K | 3000.0M | $9.8M | $117.0M |
Save your results
Get your cost estimate sent to your inbox. We'll also send tips on how to reduce your AI spending.
No spam. Unsubscribe any time.
OpenAI Model Pricing
Prices are in USD per 1 million tokens. Last verified: 2026.
| Model | Input / 1M | Output / 1M | Context | Best For |
|---|---|---|---|---|
GPT-4o flagship | $2.50 | $10.00 | 128K | Best for most tasks |
GPT-4o mini fast | $0.1500 | $0.6000 | 128K | Fastest & cheapest OpenAI option |
o1 flagship | $15.00 | $60.00 | 200K | Advanced reasoning |
o1-mini standard | $3.00 | $12.00 | 128K | Fast reasoning model |
GPT-3.5 Turbo fast | $0.5000 | $1.50 | 16K | Legacy fast model |
GPT-4o
Best for most tasks
GPT-4o mini
Fastest & cheapest OpenAI option
o1
Advanced reasoning
o1-mini
Fast reasoning model
GPT-3.5 Turbo
Legacy fast model
Always confirm at the provider's official pricing page before making business decisions.
How OpenAI pricing works
OpenAI pricing is built on a usage-based model where you pay for what you use. They measure this usage in "tokens," which are the basic building blocks of text processed by the model.
When you send a request to an OpenAI model, you are billed for two distinct components: input tokens (the instructions and context you provide) and output tokens (the text generated by the AI in response). Each component has its own price point, typically with output tokens costing significantly more than input tokens.
The flagship model, GPT-4o, provides the highest level of intelligence and versatility, making it the best choice for complex reasoning tasks. For high-volume applications where cost efficiency is paramount, GPT-4o mini offers a highly capable alternative at a fraction of the price, often delivering results that are indistinguishable for simpler or more structured tasks.
