No. ChatGPT Plus and ChatGPT Pro do not include OpenAI API access or API credits. OpenAI runs ChatGPT subscriptions and the API platform as separate billing systems (opens in new tab). Paying for ChatGPT does not pay for API calls, and adding API billing does not give you a ChatGPT subscription.
That distinction confuses a lot of people. It also creates a useful choice: pay a recurring fee for the consumer app, or use an API key and pay for the tokens you actually send and receive. API access can cost less for intermittent use, while a subscription can be easier to predict for heavy daily use. The honest answer depends on how you work.
This guide explains API billing, the tradeoffs, and how to get keys from OpenAI, Anthropic, and Google AI Studio.
What Are API Keys?
API keys are how developers access AI models directly. Instead of using ChatGPT's consumer interface, you connect straight to the provider's API and pay only for what you use.
Think of it like this:
- Consumer apps (ChatGPT Plus, Claude Pro, Google AI Pro): You pay a monthly subscription to use the provider's app under that plan's current features and usage limits.
- API keys: You use another client or your own software, choose a named model, and pay the API provider according to token usage. API rate limits and account spending limits still apply.
Direct API access. Add your OpenAI, Anthropic, and Google AI Studio keys. Pay them directly. No middleman markup. No subscription rent-seeking.
When API Keys Are Better Than a Subscription
1. You Pay for Usage
Consumer apps charge monthly subscriptions whether you use them once or a thousand times.
APIs bill input, output, cached, and sometimes reasoning tokens at rates that vary by model. Check the provider's current pricing before choosing a model. For light or irregular use, metered billing can be cheaper because idle months cost nothing. Heavy use can cost more than a subscription, so set a budget and inspect actual token usage instead of assuming the API always wins.
2. You Choose the Exact Model
Consumer apps can change their model picker, limits, and routing behavior. With direct API access, your request names a specific model. The response records which model handled it, making comparisons and cost tracking much clearer.
Choose any model from OpenAI, Anthropic, or Google. What you pick is what you get. No secret routing to cheaper versions.
3. The Limits Are Explicit
APIs still have rate limits, usage tiers, account limits, and model availability rules. The difference is that providers document those limits for developers, and you can usually raise them through account history or a provider request. You also control your own spending ceiling.
4. Full Technical Control
Consumer apps expose a smaller set of controls. APIs can offer additional parameters, depending on what each model supports.
With API access, you control everything:
- Temperature (creativity vs consistency)
- Max tokens per response
- Provider-specific sampling controls
- Custom system prompts
- Streaming vs batch responses
This is the same interface developers use, presented through an API client instead of a playground.
5. You Can Use Multiple Models Without Multiple Subscriptions
Want to try GPT-5 for creative writing, Claude for analysis, and Gemini for fast responses?
- Consumer route: Monthly subscriptions for ChatGPT Plus + Claude Pro + Google AI Pro. Multiple recurring charges.
- API route: Add keys for each provider. Pay only for tokens you actually use across all three.
Switch models mid-conversation. Compare responses. Use the best model for each task. No need to maintain multiple monthly subscriptions.
Every model from OpenAI, Anthropic, and Google in one app. Switch mid-conversation. Compare responses. Pay only for what you actually use.
But I'm Not a Developer...
You don't need to write code. You do need to create developer billing accounts, protect the keys, and understand that usage creates separate provider charges.
And once you have them, you get access to features like Face/Off Mode that generate multiple responses and pick the best one automatically.
Here's exactly how to do it:
How to Get OpenAI API Keys (Step-by-Step)
Step 1: Create an OpenAI Account
Go to https://auth.openai.com/create-account (opens in new tab)
Sign in to the OpenAI platform. Your login can use the same identity as ChatGPT, but API billing and usage remain separate from your ChatGPT subscription.
Step 2: Add a Payment Method
OpenAI requires a payment method for API access. This is pay-as-you-go billing. You're charged only for what you use.
Go to https://platform.openai.com/settings/organization/billing (opens in new tab)
Add a credit card. Set a spending limit if you're worried about costs (e.g., $10/month max).
Step 3: Create an API Key
Go to https://platform.openai.com/settings/organization/api-keys (opens in new tab)
Click "Create new secret key."
Give it a name (e.g., "My iPhone").
Copy the key. You'll only see it once. Store it somewhere safe temporarily (you'll enter it in the app next).
How to Get Anthropic API Keys (Step-by-Step)
Step 1: Create an Anthropic Account
Go to https://console.anthropic.com/login (opens in new tab)
Sign up with email. This gets you access to the Anthropic Console and API.
Step 2: Add a Payment Method
Like OpenAI, Anthropic and Google AI Studio use pay-as-you-go billing.
Go to https://console.anthropic.com/settings/billing (opens in new tab)
Add a credit card. Anthropic might give you some free credits to test with.
Step 3: Create an API Key
Go to https://console.anthropic.com/settings/keys (opens in new tab)
Click "Create Key."
Copy the key. Store it somewhere safe temporarily.
How to Get Google AI Studio API Keys (Step-by-Step)
Step 1: Create a Google Account
Go to https://aistudio.google.com/ (opens in new tab)
Sign in with your Google account. This gives you access to Google AI Studio and the Gemini API.
Step 2: Enable Billing
Google AI Studio requires billing for API access beyond free tier limits.
Go to https://console.cloud.google.com/billing (opens in new tab)
Set up a billing account. Google uses pay-as-you-go pricing. Compare the current model rates and free-tier terms in Google AI Studio before choosing a model.
Step 3: Create an API Key
Go to https://aistudio.google.com/apikey (opens in new tab)
Click "Get API key" or "Create API key."
Copy the key. Store it somewhere safe temporarily.
Now What? How Do I Actually Use These Keys?
You need an app that supports direct API access. You've got a few options:
- OpenAI Playground / Claude Console / Google AI Studio: Free browser tools. Decent for testing, but browser-only. No mobile app. No cross-device sync.
- Build your own: If you're a developer, you can build a custom interface. Most people don't want to do this.
- Use Cumbersome: Native iPhone, iPad, and Mac app that supports your OpenAI, Anthropic, and Google AI Studio API keys directly.
Cumbersome is built specifically for this use case. It's 100% free. You add your keys, and the app connects directly to the providers on your behalf.
Settings → API Keys. Add your OpenAI, Anthropic, and Google AI Studio keys. The app connects directly to their APIs. Your conversations never pass through our servers. We don't see your data. We're completely out of the money flow.
What About Privacy?
API keys do not make provider processing disappear. OpenAI, Anthropic, or Google still receives the prompts you send to its API, and each provider's current retention terms apply.
With API keys in Cumbersome: Your device connects directly to OpenAI, Anthropic, and Google AI Studio. Your conversations sync via your personal iCloud account, not our servers. We cannot see your conversation history because it never passes through our infrastructure.
This is architectural, not a privacy policy promise. We don't have a backend database. We can't see your conversations even if we wanted to.
iPhone, iPad, Mac. Your conversations sync via your personal iCloud account. API keys sync via iCloud Keychain. We never see your data. Privacy by architecture, not by policy.
But What If I Use It A LOT?
Heavy API use can cost more than a consumer subscription. That is the tradeoff. API billing gives you model choice and usage-based pricing, but it does not promise a lower bill.
Set provider spending limits where available, choose less expensive models for routine work, and review token usage before committing to high-volume workflows.
What About API Rate Limits?
APIs do have rate limits. These cap how many requests you can make per minute or how many tokens you can process per day. New accounts start with aggressive limits.
Providers assign API limits by model, account tier, and usage history. Some tiers increase automatically; others require a request or different account terms. Check the provider dashboard instead of assuming a new key supports unlimited traffic.
Subscription or API: Which Should You Choose?
Choose a consumer subscription when you want the provider's complete app, bundled features, simple billing, and no API-key management.
Choose API access when you want to name the model, compare providers, use a third-party client, control parameters, or pay only for measured usage. Many people use both because the products solve different problems.
Who Should Use API Keys Instead of Consumer Apps?
API access is a good fit for:
- Intermittent users who prefer measured usage to another recurring subscription
- Power users who need explicit controls over models, parameters, and context
- People who want fewer intermediaries between their device and the API provider
- Multi-model users who want to try different AI models without multiple subscriptions
- Anyone who likes transparency over black-box interfaces
Stick with a consumer app when:
- You do not want to manage keys, provider billing, or token costs
- You rely on bundled features that are not exposed through the API
- Your usage makes a predictable subscription the better value
Try It
Get Cumbersome for iPhone, iPad, or Mac. The app is free. Add your API keys, choose a model, and see the token usage behind every request.
Bless up! 🙏✨