Skip to main content

​
Pricing

For full pricing per model see our dedicated pricing page with all models listed. We do not charge any minimum fee per query, nor do we charge any fee on deposits. Pricing examples and model IDs follow the same canonical-ID policy as this documentation: use exact IDs returned by GET /api/v1/models, and do not rely on hidden/internal aliases. If you are (potentially) going to be a large user of our API reach out to us at support@nano-gpt.com or our Discord for a discount.

​
Look up a recent request charge

Request Billing returns the original recorded primary charge and token usage using the response’s X-Request-ID and the same inference API key. It is available for 24 hours from the charge’s accounting timestamp. The exact cost is a decimal string in USD or XNO; refunds and separately billed extras are excluded. For aggregate spend and refund totals, use Usage.