Skip to main content

Billing and Usage Tracking

Outter uses a pay-as-you-go billing model with possible monthly quotas. Every API call you make is tracked for billing purposes. Billing is often measured in AI compute units like tokens for language models, or by count of requests for simpler endpoints. The system emits usage events for each request (with details like tokens or operations) which the Billing Service consumes. This usage is aggregated per tenant (your account) for billing.

Key points:

  • Token-Based Billing: For endpoints that use language models (generation, chat, etc.), billing is typically based on the number of tokens processed. A token is roughly 4 characters of text; this is the unit used by models like GPT. If you generate a 100-word response, that might be ~150 tokens, for example. Outter will count the input tokens plus output tokens for the request.
  • Operation-Based Billing: For other services like recommendations or data queries, billing might be per request or per item processed. For example, one recommendation call returning 5 items might count as 5 recommendation operations or some fixed credit per call.
  • Usage Records: You can usually view your usage in the Outter dashboard or via an API. Outter’s multi-tenant database keeps usage logs for each tenant.
  • Costs: Each model or operation might have a different cost. For example, using GPT-o3 might cost more per 1k tokens than GPT-4o. Outter will translate the raw usage into cost based on a pricing table. If on a paid plan, you’ll be charged for the total usage (often monthly). If on a free tier with quota, you won’t be charged up to the quota; after that, you may incur overage fees or the service might throttle depending on the plan.

Transparency: Outter aims to be transparent in usage. API responses for AI calls include usage metadata. For instance, after a content generation, the JSON response might include something like:

{
  "result": "generated text...",
  "usage": { "promptTokens": 50, "responseTokens": 120, "totalTokens": 170 }
}

This helps you understand how many tokens were used for that call. You can sum these to track your own usage on the fly. Additionally, the Outter dashboard will show your cumulative usage and cost in real time.