ChatGPT API Pricing in 2026
ChatGPT Plus is a $20/month flat fee for the chat app. The OpenAI API is pay-per-token — a separate product with separate billing at platform.openai.com. RunAPI mirrors every GPT model at half the official rate, same endpoint, same output, 50% less on your invoice.
What does the ChatGPT API cost right now?
ChatGPT has two completely separate products with different billing. The subscriptions (Plus, Pro, Team) are flat monthly fees for the chat app — they do not include API credits. The API charges per million tokens and gives programmatic access from code. If you paid for Plus thinking it gave you API access, it doesn't — that's a separate account at platform.openai.com. The cards below cover both.
ChatGPT Plus is $20/month for the web and mobile chat app. No API access and no per-token billing.
GPT-5.4 at $1.25/M input and $7.50/M output through RunAPI. Note that output tokens cost 6× more than input — keep responses short to control spend. Official rate is $2.50/$15.
GPT-5.4-mini at $0.13/M input and $0.50/M output through RunAPI. Official rate is $0.25/$1.
The API runs in Claude Code, Cursor, and Codex through an OpenAI-compatible endpoint. Subscriptions do not.
What is the difference between ChatGPT Plus, Pro, Team, and the API?
The subscriptions are for people who chat in the app. The API is for developers who call the model from code — it has no message cap; you pay per token and scale as needed. Many developers are shocked that the OpenAI website makes this so hard to find. The table compares all four so you can match the plan to how you actually use ChatGPT.
| Plan | Price | What you get | Best for |
|---|---|---|---|
| ChatGPT Plus | $20/mo | Chat app, GPT-5.4 access, higher limits than free | Individuals using the chat interface |
| ChatGPT Pro | $200/mo | Chat app, highest limits, Pro reasoning mode | Power users running long reasoning tasks |
| ChatGPT Team | $25–$30/user/mo | Chat app for teams, shared workspace, admin console | Small teams sharing the chat app |
| OpenAI API | Pay-per-token | Programmatic access, all models, no chat app | Developers, agents, CI pipelines, editors |
The API and the subscriptions are billed separately. Paying for Plus does not include API credits, and API usage does not unlock the chat app.
How much does each GPT model cost per million tokens?
The table below shows official OpenAI pricing alongside RunAPI pricing. RunAPI applies a flat 50% discount across all GPT models. No volume commits, no subscriptions.
| Model | Official input /M | Official output /M | RunAPI input /M | RunAPI output /M | Context window |
|---|---|---|---|---|---|
| GPT-5.5 | $3.00 | $18.00 | $1.50 | $9.00 | 256K |
| GPT-5.4 | $2.50 | $15.00 | $1.25 | $7.50 | 256K |
| GPT-5.4-mini | $0.25 | $1.00 | $0.13 | $0.50 | 256K |
| GPT-5.2 | $2.50 | $15.00 | $1.25 | $7.50 | 200K |
| GPT-5.3-codex | $2.50 | $15.00 | $1.25 | $7.50 | 256K |
What does the ChatGPT API actually cost for real workloads?
Token costs look abstract until you attach them to real tasks. Below are five common developer workloads with estimated monthly costs at two usage levels through RunAPI. Output tokens are priced 4–6× higher than input tokens, so prompts that generate long responses cost more than they look — these estimates account for that.
| Workload | Model | Light use (~50 tasks/day) | Heavy use (~200 tasks/day) | Monthly saving vs official |
|---|---|---|---|---|
| Coding session in Cursor | GPT-5.4 | $38/mo | $150/mo | $38–$150 |
| Codex agent runs | GPT-5.3-codex | $50/mo | $200/mo | $50–$200 |
| Support chatbot | GPT-5.4-mini | $6/mo | $24/mo | $6–$24 |
| Content generation pipeline | GPT-5.4 | $28/mo | $112/mo | $28–$112 |
| Multi-agent orchestrator | GPT-5.5 | $120/mo | $480/mo | $120–$480 |
How to access the ChatGPT API through RunAPI
Create a RunAPI account
Sign up at runapi.ai. No credit card required for the free tier.
Copy your API key
Go to Dashboard → API Keys. Create a key and save it — you will use this as your OpenAI API key.
Point your SDK to RunAPI
Set the base URL to https://runapi.ai/v1 and use your RunAPI API key. Any OpenAI-compatible client works — Python, Node.js, Go, Ruby, or curl.
Start making requests
Use gpt-5.4, gpt-5.4-mini, or any GPT model ID in the model parameter. RunAPI handles routing and billing at 50% of the official rate.
ChatGPT API Pricing FAQ
Is ChatGPT Plus the same as the ChatGPT API?
No. ChatGPT Plus is a $20/month subscription for the chat app on web and mobile. The API is a separate pay-per-token product for developers. Paying for Plus does not include API credits, and API usage does not unlock the chat app.
How much does GPT-5.4 cost through the API?
Official OpenAI pricing for GPT-5.4 is $2.50/M input and $15/M output. Through RunAPI the same model costs $1.25/M input and $7.50/M output — a flat 50% discount with no subscription or volume commitment. You only pay for the tokens each request actually uses.
Can I use the ChatGPT API in Cursor or Codex?
Yes. Both Cursor and Codex accept an OpenAI-compatible base URL. Point them at runapi.ai/v1 with your RunAPI key and select a GPT model ID. The subscriptions cannot be used this way — only the API gives the programmatic access these editors need.
Why is RunAPI 50% cheaper than the official API?
RunAPI negotiates volume pricing with model providers and passes the savings to developers. There is no quality difference — same models, same OpenAI-compatible API, same output. You only change the base URL and key, and your existing client code keeps working unchanged.
What is the cheapest GPT model for the API?
GPT-5.4-mini is the cheapest, at $0.13/M input and $0.50/M output through RunAPI. It suits high-volume tasks like classification, support chat, and extraction where the larger models would be overkill. For complex reasoning, step up to GPT-5.4 or GPT-5.5 instead.
If I run an n8n bot that creates a few outputs a week, how much will the API cost?
For a low-intensity automation — say an n8n bot running a few tasks a week — the API costs very little. Three short tasks a day at roughly 1,000 tokens each comes to under $1/month on GPT-5.4-mini through RunAPI. Costs only spike if the tasks process large documents or run hundreds of times a day.
Do they allow you to use it free for a certain amount of time?
RunAPI has a free tier with no credit card required. You get a set of free credits to test requests before you add a balance. Once those run out, you fund your account and pay only for what you use — there are no monthly minimums.
How does billing work?
Pay-as-you-go. You fund your account with a balance, and each API call deducts the token cost. No subscriptions, no invoices, no contracts. You can monitor spend per key from the dashboard.
Explore More
GPT Models on RunAPI
Browse all GPT variants — GPT-5.4, GPT-5.5, and more with per-call pricing.
OpenAI API Pricing
Full OpenAI model lineup — GPT-5.4, o3, and more at 50% off.
Claude API Pricing
Claude Opus, Sonnet, and Haiku pricing at 50% off Anthropic rates.
All Model Pricing
Compare 220+ models across text, image, video, and audio.
Start using GPT at half price.
Create a free RunAPI account, get your API key, and start calling GPT-5.4, GPT-5.4-mini, or any GPT model at 50% off official OpenAI pricing.