orkaAI Support Cost Calculator
Estimate model usage costs for customer support using your own token prices. Enter the tokens billed over an entire conversation, including any calls that search knowledge or draft a follow-up.
Worked example
At $1 per million input tokens and $4 per million output tokens, 2,400 input tokens cost $0.0024 and 300 output tokens cost $0.0012. That is $0.0036 per conversation, or $36 for 10,000 conversations.
How the calculation works
Multiply billed input tokens per conversation by the input rate per million tokens and divide by one million. Repeat for output tokens at the output rate. Add both amounts for unit cost and multiply by monthly AI-assisted conversations. Prices in the example are illustrative USD rates.
Frequently asked questions
Where should I get my model’s token prices?
Use the rates for your model and billing tier from the provider’s pricing page or your latest invoice. Enter each price in US dollars per one million tokens.
Why does a long conversation use more input tokens?
Many applications send prior messages, instructions and retrieved knowledge back to the model on subsequent turns. Sum billed input across those calls or use the mean total from a representative usage export.
How do caching discounts and other charges affect this estimate?
For cached and uncached input, enter their effective blended input rate from a representative billing period. Add search fees, storage, platform subscriptions and taxes separately when planning an overall budget.