Skip to content
BrainRoad BrainRoad
Documentation Menu

API Costs Explained

How AI model usage costs work: trial credit first, then bring your own API key. BrainRoad never marks up token costs.

On this page

Two Separate Costs

Using BrainRoad involves two independent costs:

  1. Your BrainRoad subscription pays for your private Business Brain and hosted helper workspace: storage, infrastructure, and the dashboard. Fixed price (Plans & Pricing).
  2. AI model usage pays for the tokens your helper consumes when it reads, thinks, and drafts. Variable, and billed by the AI provider, not by BrainRoad.

BrainRoad never touches your model spend. There is no markup, no middleman fee, and no surcharge on token usage. Pro does not include model usage.

During the Trial: BrainRoad Covers It

During the 30-day trial, BrainRoad provides a trial API key (through OpenRouter) with a small credit balance, currently $5.00, so you can evaluate without setting up a provider account. The dashboard warns you when the credit runs low. See What Happens During Your Trial.

After the Trial Credit: Bring Your Own Key

Once the trial credit is used (or any time earlier, if you prefer), you add your own API key and the provider bills you directly. BrainRoad supports keys from Anthropic, OpenAI, Google, OpenRouter, xAI, and more — see Choosing a Provider for the full list.

To add a key, open the Helper page in the dashboard and use the key setup there (Use My Own API Key, or Add Key Manually if you already have one). Your helper switches to your key immediately.

What Model Usage Costs

Costs depend entirely on which model your helper uses and how much work it does. Longer conversations, bigger documents, and more frequent tasks all mean more tokens. Check your provider’s current pricing page for per-token rates; they change often enough that we do not print them here.

Managing Model Costs

  • Choose an efficient model. Smaller, cheaper models handle many routine tasks well. You can configure which model your helper uses in its settings.
  • Set spending limits at the provider. Most providers let you cap monthly API spend on your account. This is the most reliable way to prevent surprise bills.
  • Watch usage in the dashboard. The Usage page (dashboard sidebar) shows token usage and links through to your helper runtime’s detailed per-model usage view. The Helper page also shows an at-a-glance cost summary. For exact billing, your provider’s own usage dashboard is the source of truth.
  • Start small. Begin with one provider and a conservative pattern, then scale once you understand your helper’s typical consumption.