Benjamin Satzger d18d16e0ba Introduce billing for inference endpoints ...

This commit introduces a new `InferenceTokens` billing resource type.
Billing rates are added that define how much a user is billed for each
inference token used. Billing depends on the amount of tokens and the
type of model used for inference.

2024-10-16 13:56:06 +02:00

1.7 KiB

Raw Permalink Blame History

View Raw