Pricing & Billing

Docs-depth version of the pricing page. What counts as a request, how cost is computed, and billing mechanics.

Billing tiers at a glance

No seats and no monthly minimums. Model usage is billed by your provider on your own keys: Spineforge never marks up tokens.

What counts as a request?

One call through the gateway to a model endpoint (for example chat completions). A streamed response counts once. Requests the gateway rejects (a revoked key, an exceeded budget cap, or a scope violation) are not billed.

What counts as an agent?

Any unique gateway key that has made at least one request in the billing period. Keys you've created but never used are not counted. Test and CI keys count if they send traffic.

Free tier limit

On Free, requests pause once the monthly allowance of 10,000 is used. Upgrade to Pay-as-you-go at any time to continue without interruption.

How cost is computed: the formula

This is the cost of the model call shown on your dashboard (the figure your provider charges), computed for each gateway request:

cost_usd = (prompt_tokens    / 1000) × cost_per_1k_prompt_tokens
         + (completion_tokens / 1000) × cost_per_1k_completion_tokens

# Stored as NUMERIC(14,8): precision to sub-cent

Cost is computed once, at write time, using the rate card in effect at that moment. It is stored on the request row with the rate_card_id that produced it and never recomputed. See Metrics Reference for the full step-by-step pipeline.

Billing period and invoicing

• Billing periods are calendar months (UTC).
• Pay-as-you-go: invoiced monthly in arrears for requests above the free allowance.
• Enterprise: invoiced per custom agreement.
• Free tier: no payment method required.

Data retention

Next steps

• Pricing page: interactive estimator and tier comparison
• Metrics Reference: how every cost figure is computed