Inference pricing

Your organization receives a one-time free inference credit. It does not reset each month. After you use it, buy AI credits before making more requests.

Free credit

Inference credits cover model requests routed through Jotsu. This includes direct API calls, AI features in your applications, and interactive Codex sessions using Jotsu inference.

The free credit is the total free Jotsu inference available to your organization. It remains available until you use it, but it does not renew or become cash.

Buy more credits

When the free credit reaches zero, Jotsu blocks new funded requests. An organization owner must add a payment method and choose an AI credit package before usage can continue.

Adding a card does not charge it by itself. The charge happens when the owner confirms a credit purchase. Purchased credits are not a subscription and do not reset each month.

AI credits may be shown as AI tokens in the product. They represent prepaid usage value, not a fixed number of raw model input or output tokens.

How credits are used

Jotsu calculates each request from the model's input and output token prices. Purchased credit usage is:

underlying model cost × 1.15

For example, a request with an underlying model cost of $1.00 uses $1.15 of purchased credit.

Input, cached input, and output tokens may have different prices. Long-context requests may also use a different price tier.

The balance stops at zero. Jotsu does not run postpaid inference, create an overage balance, or charge the saved card automatically.

Codex with connected credentials

Interactive Codex requests made through credentials connected by your organization do not consume the free grant or purchased Jotsu AI credits. The connected provider or subscription controls their cost and limits.

Application inference and other Jotsu-managed model calls continue to use the organization's Jotsu AI balance even when interactive Codex uses connected credentials.

See Codex model access to compare the two access methods.