Short answer
“ChatGPT 4.0” is not the model name used in OpenAI’s current pricing pages. If you mean GPT-4o, note the important distinction: GPT-4o was retired from ChatGPT on February 13, 2026, but OpenAI says it remains available through the API. OpenAI’s API model page marks GPT-4o as deprecated. So a ChatGPT subscription is not a way to select GPT-4o in ChatGPT today; GPT-4o’s listed API rate is relevant to API use, particularly existing integrations. (OpenAI Help Center)
For individual ChatGPT access, the listed prices are Free at $0/month, Go at $8/month, Plus at $20/month, and Pro starting at $100/month. For API use, GPT-4o is listed at $2.50 per million input tokens and $10 per million output tokens. These are separate billing routes: ChatGPT plan fees do not ordinarily pay API charges. (ChatGPT plans and pricing)
ChatGPT plan prices
| Individual plan | Listed price | What to know |
|---|---|---|
| Free | $0/month | Limited access to some features and tools. |
| Go | $8/month | Higher limits for eligible features than Free; ads may appear for eligible users. |
| Plus | $20/month | Higher model limits and broader model options than Free; model availability and usage limits can change. |
| Pro | From $100/month | OpenAI lists monthly tiers: Pro 100 at $100, Pro 200 at $200, and Pro 500 at $500. |
These are USD prices shown on OpenAI’s pricing and plan pages. Go subscriptions are billed automatically monthly; Plus is $20/month, billed monthly, and OpenAI says annual billing is not available for Plus. Pro tiers are also monthly, with no annual billing. OpenAI notes that prices and available options can vary by country or account, so check the country selector and checkout before subscribing. (ChatGPT pricing; ChatGPT Go billing; ChatGPT Plus billing; ChatGPT Pro tiers)
If you are asking specifically for GPT-4o in ChatGPT: there is no current individual ChatGPT plan price that buys access to that retired model in ChatGPT. A plan provides ChatGPT access under its current features and model availability, not continued access to every older model. (OpenAI Help Center)
GPT-4o API pricing and a sample budget
OpenAI lists these standard GPT-4o API rates, in US dollars per 1 million tokens:
| Token type | GPT-4o API rate |
|---|---|
| Input tokens | $2.50 per 1 million |
| Output tokens | $10.00 per 1 million |
The API page marks GPT-4o as deprecated, so these rates are most useful when estimating an existing GPT-4o integration. For a new production integration, OpenAI recommends evaluating GPT-6 Luna as a starting point for GPT-4.x migrations and comparing quality, latency, and cost on representative requests. (GPT-4o API model page)
Worked example: 1 million input tokens and 100,000 output tokens
Using the listed GPT-4o rates:
- Input: 1,000,000 ÷ 1,000,000 × $2.50 = $2.50
- Output: 100,000 ÷ 1,000,000 × $10 = $1.00
- Estimated total: $3.50
This is an illustrative text-token calculation, not a measured workload or a quote. It excludes any applicable tool, storage, or modality-specific charges, and assumes the stated token volumes are billed at these standard rates. Actual API cost depends on the model, features, and usage. (GPT-4o API model page)
As one current alternative to evaluate, OpenAI lists GPT-6 Luna at $0.10 per million input tokens and $0.50 per million output tokens under standard pricing. At the same illustrative volumes, the token charge would be $0.15: $0.10 for input plus $0.05 for output. That arithmetic does not establish that the models perform equally on your task; test representative examples before switching production traffic. (OpenAI API pricing)
Subscription or API: which cost applies?
- Choose a ChatGPT plan if you want to use ChatGPT directly through its app or website and the current plan’s models and limits meet your needs.
- Use API pricing if you are building an application or automating model requests. Estimate cost from your expected input and output tokens, then account for any tools or modalities your application uses.
- Do not treat Plus or Pro as API billing. OpenAI says ChatGPT and API billing are separate. Eligible Pro subscribers may claim API credits as a separate benefit, subject to eligibility and API terms; those credits are not the same as ChatGPT usage allowance. (OpenAI billing help)
For a new GPT-4.x production workload, OpenAI advises testing a recommended migration candidate on representative requests and comparing quality, latency, and cost before shifting traffic. Without your request volume, token counts, and modality mix, a reliable monthly API budget cannot be calculated. (OpenAI migration guidance)
Sources
Retiring GPT-4o and other ChatGPT models