How Do You Get an API Key for OpenAI, Anthropic, Gemini, Mistral, Groq and Perplexity? Quick Guide for 2026

To get an API key for a major AI model, sign in to the provider’s developer console, enable billing or confirm free-tier access, then create a key in the API keys section. This guide shows the exact path for OpenAI, Anthropic, Gemini, Perplexity, Mistral, and Groq.

What this guide covers:

• An API key is not the same as a ChatGPT, Claude, or Gemini account: the chat products and the developer APIs are separate things built by the same companies.
• OpenAI, Anthropic, and Perplexity require a payment method before your key will work; Google Gemini and Groq both offer a free tier with no credit card required.
• Most providers show your API key only once at creation. Copy it immediately and store it securely.
• API pricing is usage-based and pay-as-you-go; for most teams running structured prompts, costs are significantly lower than a flat monthly chat subscription.

What Is the Difference Between ChatGPT, Claude, and Gemini and Their Underlying APIs?

Most major commercial AI models are available in two ways: as a chat product with a user interface and as an API that developers use to build their own applications. The company behind each model offers both, but they are separate products with separate accounts, separate billing, and separate credentials. Understanding this distinction is the starting point for getting and using an API key.

The pattern is the same across every provider: the company builds the model, the chat product is one finished application on top of it, and the API is direct access to the model itself — without any interface in between.

‍

What Is the Difference Between ChatGPT and the OpenAI API?

This is the question that trips up most people getting started. ChatGPT and the OpenAI API are two separate products made by the same company.

OpenAI is the company. ChatGPT is the consumer chat product that OpenAI builds and maintains. The two are related but not the same. When you open chat.openai.com and have a conversation, you are using ChatGPT — a finished product with a user interface, conversation history, and built-in tools like image generation and file uploads. OpenAI's APIs serve a broader developer audience, while ChatGPT caters to end users of conversational AI.

The OpenAI API gives you direct access to the underlying models — from fast, non-reasoning models like GPT-4o to more advanced, reasoning-capable models such as the latest GPT-5.4 — without the chat interface. You send a prompt, you receive a response, and you decide what happens with it.

When you encounter a tool that claims to be "powered by GPT," you are using OpenAI's technology but not ChatGPT itself. Those tools connect to the GPT models through the OpenAI API,  a direct, programmatic line to the models that bypasses the chat interface entirely. ChatGPT uses flat‑rate subscription plans with usage caps that vary by tier, while the API is priced on a metered, pay‑per‑token basis — you pay for what you actually use.

‍

What Is the Difference Between Anthropic and Claude?

Anthropic is the company that created and maintains the Claude model family. Claude.ai is Anthropic's chat product — a productivity-focused interface with Projects, document analysis, and long-context capabilities. A Claude Pro subscription unlocks higher usage within that interface. It does not unlock API access. This is the most common source of confusion for people who have been using Claude.ai and want to start building with it.

The Anthropic developer console at console.anthropic.com is an entirely separate platform with its own account, its own billing, and its own API credentials. A Claude Pro subscriber who wants to use the API still needs to complete the full developer setup — including adding a payment method to the console — before any API call will work.

‍

What Is the Difference Between Google and Gemini?

Google is the company. Gemini is both the name of Google's AI model family and the name of their consumer chat product at gemini.google.com — which is integrated into Google Workspace apps like Docs, Gmail, and Drive, and available as part of Google One subscriptions.

The Gemini API is accessed through a completely different platform: Google AI Studio at aistudio.google.com. A standard Google account is enough to log in, and a Gemini Advanced subscription has no bearing on API access — the two are entirely independent. Currently Google AI Studio offers one of the most generous ongoing free tiers among major providers: you can make Gemini API calls on the free tier with no credit card required, subject to per‑minute and per‑day limits and model restrictions.  API keys are generated within AI Studio and carry their own usage quotas, separate from anything tied to a Gemini chat subscription

‍

Popular AI Chat Products and Their APIs

‍

When you use a tool or application that says it is "powered by GPT" or "powered by Claude," you are using OpenAI's or Anthropic's technology but not the chat products themselves. Those tools have an API key behind them. That is exactly what this guide helps you get.

‍

What Are the Benefits of Using an AI API Instead of a Chat Product?

Using the API rather than a chat interface gives you three things a chat subscription does not.

• You connecting multiple providers simultaneously, making multi-testing and comparison easier in an interface that allows it, such as Promptitude (and even other providers).
• As a connector, it ensures transparent use since the client has control of their own account with the provider, thus preventing security vulnerabilities in data handling. It also prevents providers from training their models through external means rather than via the provider's chat.
• You pay only for what you use. Instead of paying a flat fee like a chat subscription, you pay for input and output tokens as you use them. For structured, repeatable prompts — as opposed to open-ended conversations — API costs per task are typically far lower than a monthly seat.

‍

How Do You Get an OpenAI API Key?

Go to platform.openai.com and sign up or log in. OpenAI requires you to verify your email address and may request a phone number during registration. Once logged in:

1. Open the Billing page — click Billing in the left sidebar and add a payment method. OpenAI billing requirements can change. In many cases, you need to add a payment method or purchase credits before API calls will work. Free trial credits, when available, are limited-time and not guaranteed for all new accounts.
2. Go to API Keys — in the left sidebar, click API Keys.
3. Click Create new secret key — give it a descriptive name so you can identify it later.
4. Copy the key immediately — OpenAI shows your key only once. Store it securely; never commit it to a public repository.

‍

How Do You Get an Anthropic API Key?

Go to console.anthropic.com and click Build with Claude or Sign Up. Anthropic does not use passwords — it sends a one-time magic link to your email address. Click the link to verify and log in.

1. Add a payment method — click Settings → Billing and add a credit card. Anthropic’s API is usage-based and in most cases you need to enable billing or prepaid credits before you can use the key, though promotional or trial credits may occasionally be available.
2. Go to API Keys — in the left sidebar, click API Keys.
3. Click Create Key — give your key a descriptive name.
4. Copy the key immediately — Anthropic shows it only once.

‍

How Do You Get a Google Gemini API Key?

Google AI Studio offers a generous free tier that lets you call Gemini models without adding a credit card, but quotas and model access may change over time. As of 2026, free usage is limited and primarily focused on Flash/Flash-Lite models, with higher tiers and Pro models requiring billing.

1. Go to aistudio.google.com — sign in with your Google account.
2. Click Get API Key — in the left sidebar or top navigation.
3. Create API key in new project — Google creates a project automatically.
4. Copy the key — it remains visible in Google AI Studio, so you can return to it if needed.

Note: To continue using the API without interruption, restrict your key — Google is phasing out unrestricted traffic keys. In AI Studio, click the ‘Restrict to Gemini API’ button (or the current restriction option) to limit where your key can be used and keep it secure.

‍

How Do You Get a Perplexity API Key?

Visit perplexity.ai and sign up using Google, Apple, SSO, or email. Free accounts can generate API keys, but they usually do not include recurring monthly credits. Perplexity Pro subscribers often receive a monthly API credit allowance (for example, around $5), though the exact amount and terms can change, so always confirm in your account billing settings.

1. Go to your Perplexity account settings — click your profile icon in the top right.
2. Open the API section — navigate to Settings → API.
3. Click Generate API key — name your key and confirm.
4. Copy the key immediately — store it securely.

‍

How Do You Get a Mistral API Key?

1. Go to console.mistral.ai — create an account or log in.
2. Add a payment method — Mistral uses usage-based pricing; you need billing set up before your key is active. If Mistral offers any trial credits or promos, they’ll appear in your console; do not assume a permanent free tier.
3. Go to API Keys — the Mistral API key URL is console.mistral.ai/api-keys.
4. Click Create new key — give it a name and set an optional expiry date.
5. Copy the key immediately — Mistral shows your key only once.

‍

How Do You Get a Groq API Key?

Groq offers a free tier account: you can sign up at console.groq.com, create an API key with no credit card, and use their hosted models within rate limits. The free tier is capped by per-minute or per-day limits, so it’s best for experimentation rather than heavy production use.

1. Go to console.groq.com — sign up or log in.
2. Go to API Keys — click the API Keys section in the console.
3. Click Create API Key — give it a descriptive name.
4. Copy the key immediately — Groq will not show the key again once you leave the page.

How Do You Use Your API Keys in Promptitude?

If you use Promptitude, you can store each provider key and reuse it across prompts. Go to Settings → API Keys, select the provider, paste your key, and save.  That lets your team reuse the same provider setup across prompts without re-entering credentials for each workflow. From that point, every prompt in your shared library can use that provider's models — switchable from a dropdown, testable side by side, and deployable via REST API without any code changes.

Your organization's content storage connects directly to those prompts, so outputs are shaped by your own knowledge base rather than starting from scratch each time.

Ready to connect your API keys? Add your first key to Promptitude →

‍

Quick Reference: Where to Find Each API Key

Free tier details and quotas change frequently. Always confirm current limits in each provider's console or pricing page before building on the assumption of continued free access.

‍