AI Prompt Cost Calculator
Calculate the cost of an AI API call based on input/output tokens and model pricing.
Did this tool work for you?
How to use this calculator
AI providers charge separately for input (prompt) tokens and output (completion) tokens, priced per million tokens. Total cost is the sum of both.
- 1
Select the AI model you are using, or choose "Custom pricing" to enter your own rates.
- 2
Enter the number of input (prompt) tokens and output (completion) tokens per call.
- 3
Enter how many API calls you make per day to see daily, monthly, and annual cost projections.
Frequently asked questions
What formula does the prompt cost calculator use?
Cost per call = input tokens ÷ 1,000,000 × the input rate + output tokens ÷ 1,000,000 × the output rate, with both rates in dollars per million tokens. Daily cost = cost per call × calls per day, monthly = daily × 30, annual = daily × 365.
Where do the model prices come from?
They are a fixed table inside the tool, listed in the dropdown labels: for example $5 input and $15 output per million tokens for the GPT-4o entry. They are figures built into this tool, not live quotes. Vendors change prices and retire models, so check the provider’s current pricing page before budgeting.
What if my model is not in the dropdown or the price differs?
Choose Custom pricing and type the input and output price per million tokens. The two custom price fields are ignored for every other selection, and the default custom values are $3 and $15.
Why are input and output costs shown separately?
Providers bill them at different rates. In every entry of this tool’s built-in table the output rate is higher than the input rate, so a response-heavy workload costs proportionally more than the token total alone suggests.
Why do small costs show so many decimals?
Per-call amounts are often fractions of a cent, so values below $0.01 are shown with six decimals and larger ones with four. Daily, monthly and annual figures are rounded to two decimals.
How do I get the token counts?
The calculator does not tokenize text. Take the usage figures returned by the API for real calls, or use the provider’s tokenizer, and average them. Remember that input includes the system prompt, retrieved context and any conversation history you resend.
Is the monthly figure based on calendar months?
No, it is daily cost × 30 and the annual figure is daily cost × 365. Real invoices follow calendar months, so the monthly number is an approximation.
What does the estimate leave out?
It applies one flat rate to every token. Cached-input discounts, batch pricing, long-context surcharges, image or audio charges, reasoning tokens that are billed but not shown, retries and taxes are not modelled.
AI prompt cost calculator: price one call, then scale to a day, month and year
From tokens to dollars
A call’s cost is input tokens ÷ 1,000,000 × the input rate plus output tokens ÷ 1,000,000 × the output rate. Rates are dollars per million tokens. Daily cost multiplies one call by calls per day, monthly multiplies daily by 30 and annual multiplies daily by 365. Small results are shown with extra decimals, six places when a value is below one cent, because per-call figures are so often fractions of a cent. The total tokens line simply adds input and output tokens for one call.
The model table built into this tool
The figures below are stored in the code and shown in the dropdown labels. They are not live quotes. Vendors revise prices and retire models, so check the provider’s current pricing page and use the Custom option when your rate differs. The two custom price fields are ignored unless Custom pricing is selected.
| Dropdown entry | Input | Output |
|---|---|---|
| GPT-4o | $5 | $15 |
| GPT-3.5 Turbo | $0.5 | $1.5 |
| Claude 3.5 Sonnet | $3 | $15 |
| Claude 3 Haiku | $0.25 | $1.25 |
| Gemini 1.5 Pro | $3.5 | $10.5 |
| Gemini 1.5 Flash | $0.075 | $0.3 |
| Custom pricing | your input | your output |
Worked examples
Prompt cost runs. The columns show Run, Inputs and Result, one line per case.
| Run | Inputs | Result |
|---|---|---|
| Default | GPT-4o entry, 1,000 in, 500 out, 1,000 calls/day | $0.0125 per call ($0.005000 input, $0.007500 output); $12.50 per day; $375.00 per month; $4,562.50 per year; 1,500 tokens per call |
| High volume, small outputs | Gemini 1.5 Flash entry, 3,000 in, 400 out, 20,000 calls/day | $0.000345 per call; $6.90 per day; $207.00 per month; $2,518.50 per year; 3,400 tokens per call |
| Custom rates | $1 input / $4 output, 8,000 in, 1,000 out, 250 calls/day | $0.0120 per call ($0.008000 input, $0.004000 output); $3.00 per day; $90.00 per month; $1,095.00 per year; 9,000 tokens per call |
Where input tokens come from
Input tokens are everything you send: the system prompt, retrieved documents, conversation history and the new user message. Chat applications that resend the whole history grow the input with every turn, so a single average hides long conversations. The custom-rate run shows a large-context pattern: with 8,000 tokens in and 1,000 out, the input cost per call is $0.008000 against $0.004000 for output, even though the output rate is four times higher. Trimming context is often a better saving than shortening answers.
Limits of a single-call estimate
The calculator applies one flat rate to all tokens. It does not model cached-input discounts, batch discounts, long-context surcharges, separate prices for images or audio, or reasoning tokens that are billed as output but never displayed. Treat the daily and monthly lines as a first-pass budget, then compare them with the usage dashboard once real traffic exists. If the gap is large, the usual causes are prompts longer than the average you entered, retries, and traffic that is spikier than a flat calls-per-day figure assumes. Adjust the token counts first, because they are usually the least certain inputs, and only then revisit the rates.
Learn more from an authoritative source:
OpenAI Platform DocsWords to Tokens Converter
Convert between words, characters, tokens, and pages for AI models and content planning.
AI API Budget Calculator
Plan your monthly AI API budget based on usage volume, model selection, and request patterns.
Context Window Checker
Check if your text fits within the context window of popular AI models. Paste your prompt and see which models can handle it.
LLM Cost Comparison Calculator
Compare monthly token costs across major LLM models — GPT-4o, Claude, Gemini, and more — for your specific workload.
Embedding Cost Calculator
Calculate monthly and annual costs of generating text embeddings with OpenAI or Cohere models.
Fine-Tuning Cost Calculator
Estimate the training cost and ongoing inference cost of fine-tuning an OpenAI language model.
Results are estimates for informational purposes only and do not constitute professional financial, medical, legal, or technical advice. Read full disclaimer →