Manage OpenAI Settings

Configure your company's OpenAI API key, choose which models are available, set the max tokens used per document, and manage the uploaded user guide the assistant references.

Manage Anthropic Settings

Configure your company's Anthropic API key, choose which models are available, set the max tokens used per document, and manage the uploaded user guide the assistant references.

AI Usage & Billing

Backbone MRP uses a bring-your-own-account model: you supply your own OpenAI or Anthropic API key, and your AI provider bills you directly for the tokens each request uses. Backbone does not add any AI usage charge on top.

To help you keep track of that spend, the desktop app reports token usage and an estimated US-dollar cost right where the AI is used:

• AI order reading — the extraction results table shows the Tokens, Tokens In, Tokens Out, and Est. Cost for each document processed. See the AI Order Reading guide.
• AI Help (Ask a Question) — each answer shows the tokens used and an estimated cost for that message, and a footer keeps a running total for the session.
• AI Token Usage Report — a company-wide report of AI token usage and estimated cost over any date range, broken down by model and (for admins) by user. Open it from the Reports menu. See the AI Token Usage Report guide.

Estimated costs are computed from published per-model rates and are an estimate only — the authoritative charge always comes from your provider's own billing. When a figure begins with ~, the selected model isn't in Backbone's built-in price list and a conservative default rate was used.