How to Get an OpenAI API Key in 2026 (Step by Step + Alternative)

Getting an OpenAI API Key

An OpenAI API key takes a few minutes to create, but billing and project scope often trip people up. Here's the exact path, plus a multi-model gateway alternative when you do not want a separate provider account for every model family.

Step by Step

Go to platform.openai.com and sign up or log in.
Open Settings → API keys (or View API keys under your profile).
Click Create new secret key, name it, and copy it. It's shown only once — store it in a secrets manager or .env immediately.
Add billing: go to Settings → Billing and add a payment method / prepaid credits. Without credits, calls fail with a quota error.

API access and available spend depend on the project and account. Check the API billing overview before assuming trial or free usage.

Common Gotchas

"You exceeded your current quota" almost always means no billing/credits added, not a rate limit.
Keys are secret — never commit to Git or ship in a client app (proxy via your backend, see AI API for mobile apps).
Use project-scoped keys and separate keys per environment so you can revoke without breaking everything.

A Multi-Model Gateway Alternative

If you just want GPT in your app, you don't have to manage OpenAI billing separately. Through LinkModel, one key calls the full GPT-5.x family plus Claude, Gemini and DeepSeek — at the currently displayed LinkModel rate, price shown before each call:

curl -X POST https://api.linkmodel.ai/v1/chat/completions \
  -H "Authorization: Bearer $LINKMODEL_API_KEY" -H "Content-Type: application/json" \
  -d '{ "model": "gpt-5.6-terra", "messages": [{"role":"user","content":"Draft a release note: ..."}] }'

Switch to another currently listed GPT or Codex model by changing the model ID, then rerun schema and quality tests. Full rates are in the GPT API pricing guide. Confirm the active IDs and chat endpoint in the docs.

Which GPT to Use

Frontier work → GPT-5.6 Sol ($5/$30 at the short-context standard rate)
Balanced default → GPT-5.6 Terra ($2/$12)
High-volume tasks → GPT-5.6 Luna ($0.20/$1.20)
Coding agents → use the current Codex model documented for your API surface

Cut the Bill

Route bulk work to a cheap tier and cache your stable prompt prefix — details in how to reduce AI API costs and prompt caching.

Make One Test Call Without Exposing the Key

Follow OpenAI's developer quickstart: store the secret in an environment variable, then let the SDK or request read it at runtime. Do not paste the value into source code, screenshots, tickets, or client-side JavaScript.

export OPENAI_API_KEY="replace-with-your-secret"

After creating the key, follow OpenAI's current quickstart for the supported endpoint and model. Model availability varies by project, so use a model listed in your dashboard rather than copying an old model ID from a tutorial.

Key Security Checklist

Create separate projects and keys for development, staging, and production.
Apply project budgets and usage alerts before inviting more developers.
Keep production calls on your server; mobile and browser clients call your backend.
Rotate a key immediately if it appears in Git history or a public log.
Revoke unused keys and remove access when a team member leaves.
Log a key identifier, not the secret itself.

Quota Errors vs. Rate Limits

A quota or billing error means the project cannot spend under its current billing state. A rate-limit error means the project is sending requests faster than an allowed limit. Retrying a billing failure wastes time; use the dashboard to resolve it. For a rate limit, respect the response headers and retry with exponential backoff plus jitter.

ChatGPT and API billing are separate. Paying for a ChatGPT plan does not by itself fund API usage.

Bottom Line

Create an OpenAI key on platform.openai.com and add billing to make it work. For production, direct OpenAI access gives you the provider's native API surface; a gateway can reduce account and integration sprawl when you also need Gemini, Claude, or other models. Compare live rates and required features before choosing. See how to get an API key.

Start free with a $1 credit — no billing setup required.