A: It depends on the model/provider used. We separate text credits from media credits (images/videos), and each provider consumes a different amount based on generation cost.

In the AI model modal, at the top where you select the model, you can see exactly how many credits each model consumes. We try to be fully transparent there and give users flexibility to choose because we understand there's constant new models that perform better and we try to have them there as soon as we can.

Long term, we’d like to simplify and unify the system into a more standardized model while still keeping BYOK support, of course.