Platform Keys vs. Bring Your Own Key (BYOK)
Runtype gives you two ways to connect to AI models: built-in platform keys that work out of the box, or your own API keys from providers like OpenAI, Anthropic, Google, and xAI. Many teams start with platform keys and add their own keys as they grow.
Platform keys
Platform keys are API credentials included with every Runtype plan. They are ready to use the moment you sign up, so you can start building without creating provider accounts or adding any configuration.
Why use platform keys
Start building instantly — Create Flows and Agents right away with no setup.
Simple billing — AI model costs are included in your Runtype plan, so you only manage one bill.
Great for prototyping — Use them to explore the platform, build demos, and iterate quickly.
No external accounts needed — Skip signing up with individual AI providers.
Good to know
Each plan includes an execution allowance. You can see your current usage in the dashboard header. On Build and Trial plans, requests are gradually slowed after you reach the daily allowance, so you are not cut off.
Platform keys give you access to a curated selection of popular models from OpenAI, Anthropic, Google, xAI, and more. See Available AI models for the full list.
Because platform keys run on shared infrastructure, response times can vary slightly during peak periods.
Bring Your Own Key (BYOK)
BYOK lets you connect your own API keys from providers like OpenAI, Anthropic, Google, or xAI directly to Runtype. You get dedicated rate limits, access to every model your provider offers, and separate billing for more control.
BYOK is available on Growth, Team, Enterprise, and Platform plans. If you are on a Build or Startup plan and need BYOK, upgrade from Settings → Billing.
Why use BYOK
Dedicated rate limits — Your API keys are yours alone, so performance is more predictable.
Full model access — Use any model your provider offers, including beta, fine-tuned, or specialized models. See Available AI models for the complete list.
Cost transparency — Monitor and optimize your AI spending through your provider dashboard.
Higher throughput — A strong fit for production applications with high execution volume.
Compliance — Meet internal requirements for direct provider relationships.
Set up BYOK
Go to Settings → Providers.
Find your provider and click Configure.
If you do not have an API key yet, sign in to your provider dashboard and generate one.
Paste your API key, give it a name, and click Save API Key.
After you save a provider key, your Flows and Agents automatically prefer it over the built-in platform key for that provider. You will also see your key listed alongside the platform key, and you can choose the default key for each provider.
You can use platform keys and BYOK at the same time. For example, use your own OpenAI key for production Flows while relying on platform keys for Anthropic models you are still testing.
Compare platform keys and BYOK
Platform keys | BYOK | |
|---|---|---|
Setup | Instant and included with your plan | One-time setup that takes about five minutes |
Rate limits | Shared with fair-usage policies | Dedicated to your account |
Model access | Curated selection of popular models | All models from your provider |
Billing | Included in your Runtype plan | Separate provider billing |
Best for | Getting started, prototyping, and moderate use | Production workloads, cost control, and compliance |
Plan availability | All plans | Growth and above |
Move from platform keys to BYOK
When you are ready to add your own keys, the transition is seamless. Once you configure a provider key, Runtype automatically prioritizes it over the built-in platform key for that provider. You do not need to update your Flows or Agents.
Platform keys remain available as a fallback, so there is no disruption if you need to rotate or update your provider keys later.
Next steps
Connect AI model providers — Add your own provider keys and enable models in Runtype.
Explore available AI models — Review the supported models for platform keys and BYOK.
Create and edit Flows — Start building with the models you configured.
Review Logs — Monitor execution details, tokens, and cost after you start running Flows and Agents.