Plan the text-token allowance for an AI translation workload. Estimate input, instruction, output, and retry tokens per task, then scale the result to a batch or daily workload without confusing text tokens with generated media units.
Base tokens = source/prompt + instructions/context + expected output. Budget per task = base tokens × (1 + retry allowance). Total = budget per task × task count.
This budget measures text tokens processed around the task. It is useful for context-window planning and token-priced model costs, but it does not convert image pixels or diffusion steps into tokens.
Planning estimate only. Validate the result against production telemetry and current provider pricing before committing capacity or budget.
With 900 source tokens, 120 instruction tokens, 1000 output tokens, and a 5% retry allowance, the budget is 2,121 tokens per task.
Use recent production telemetry or a representative benchmark on the same model, hardware, input size, and quality settings.
No. It focuses on the entered operating assumptions; add fixed staffing, development, and platform overhead separately when building a full budget.
Use the observed fraction of attempts that must be repeated, and avoid adding the same retry overhead to more than one input.
A vendor maximum is a useful ceiling, but measured sustained performance is safer for capacity planning.
Input size, batching, model version, hardware, cold starts, safety checks, and traffic bursts can all change actual results.
| Input | Purpose |
|---|---|
| Workload values | Define the size and frequency of the task. |
| Performance values | Describe measured processing behavior. |
| Planning assumptions | Add utilization, retries, price, or headroom where applicable. |