01Deep dive
Meaning and practical implications
A fine-tuning job is training even when it is small. An offline batch prediction is inference even without a live user. Report their costs separately before assessing total system economics.
Training adjusts model parameters; inference uses a model’s parameters to compute an output. These are technical distinctions, not automatic accounting classifications.
A fine-tuning job is training even when it is small. An offline batch prediction is inference even without a live user. Report their costs separately before assessing total system economics.
Paste a week of traffic. Get the number that books the audit.
See what you're overpaying →