ChatGPT subscriptions and the OpenAI API are billed differently. ChatGPT Plus or Pro is a flat app subscription. The API is a developer product that charges based on usage, measured in tokens sent to and generated by the model.
Does this affect you?
Use this if you are building with the OpenAI API at platform.openai.com or using a third-party tool connected with your own API key. The API is separate from the ChatGPT app and its monthly plan.
Understand how token-based pricing works
Once tokens make sense, API bills are easier to predict.
- A token is a small piece of text, roughly three-quarters of an English word on average.
- Every request has input tokens, meaning the prompt, instructions, and conversation history you send.
- Every request also has output tokens, meaning the text the model generates back.
- Output tokens usually cost more than input tokens because generation uses more compute than reading.
- Prices are listed per million tokens, not per message, because many individual calls cost only a fraction of a cent.
- If your app sends the full conversation history each turn, every old message included in the request is billed again as input.
A short exchange with a smaller model may cost a tiny fraction of a cent. A long document processed by a more capable model can cost more. Real cost depends on model choice, input size, output length, and volume.
Check pricing for the specific model you use
Different OpenAI models can have very different costs.
- Open the official OpenAI pricing page from platform.openai.com.
- Find the exact model name your code calls.
- Check separate input and output price columns.
- Choose the least expensive model that performs the task reliably.
- Watch for extra categories such as image input, long context, cached input, or reasoning tokens if the model uses them.
Control spending and avoid surprise bills
Set limits before testing heavily.
- Open your OpenAI account billing settings.
- Set a monthly budget, hard usage limit, or alert where available.
- Review the Usage dashboard while developing.
- Cache test results locally when debugging so the same prompt is not sent repeatedly.
- Remember that failed app logic can still cost money if the API request reached OpenAI.
More control
ChatGPT Plus does not include API usage
Using ChatGPT in the app does not use your API balance, and paying for ChatGPT Plus does not make API calls free. They are separate products with separate billing.
Trial credits may expire
Some accounts receive trial credits, but they are not an ongoing free tier like ChatGPT web access. After credits expire or run out, API use requires active billing.
Test with real prompts
Run a few realistic requests, check the Usage dashboard, then multiply by your expected daily or monthly volume. That is more reliable than guessing token counts manually.
Use official pricing for current numbers
OpenAI changes prices when models and billing features evolve. Always check the live pricing page before budgeting a production project.
Sources
- OpenAI – API Pricing (2026)
- OpenAI Help Center – Difference between ChatGPT Plus and the API (2025)
- OpenAI Platform – Usage limits and billing (2026)