PLAN BEFORE YOU CALL
Model API cost calculator.
Estimate a month of text requests or image billing units using the current provider catalog. No account or API call required.
Same token volume, different base prices
Cost comparison only. Model quality, tokenization, reasoning use and latency are not assumed equal.
| Model | Input / 1M | Output / 1M | Estimated month |
|---|---|---|---|
| Google: Lyria 3 Pro Preview | $0.00 | $0.00 | $0.00 |
| Google: Gemini 2.5 Flash Lite | $0.10 | $0.40 | $2.00 |
| OpenAI: GPT-4.1 Mini | $0.40 | $1.60 | $8.00 |
| Google: Gemini 2.5 Flash | $0.30 | $2.50 | $9.25 |
How token costs are calculated
Per-request base cost = input tokens × input price per million ÷ 1,000,000 + output tokens × output price per million ÷ 1,000,000. Monthly cost multiplies this by your request count, before per-request rounding and other billing adjustments.
How image costs are calculated
For each billing category, choose one price variant and multiply its unit price by expected units. Add the categories together, then multiply by monthly requests. Do not add alternative quality variants as if they were separate charges. A provider usage receipt is the best input for forecasting repeated image work.
Why the balance reservation can be higher
A generation reserves a conservative maximum before calling the provider. Final usage settles against that reservation and releases the unused amount. This calculator estimates your chosen workload; it does not replace that admission quote.