Prompt Cost Calculator
100% PrivateAI & Prompt ToolsEstimate what a prompt would cost on pay-per-token AI APIs, compared across popular models.
Runs entirely in your browser. Your input is not uploaded to Ovatly or anywhere else.
An estimate from a fixed price table, not a live quote. Snapshot: November 2026. Prices and limits change; verify with the provider before budgeting.
- 1Choose a model from the snapshot table.
- 2Paste your prompt and click Count input tokens, or type token counts directly.
- 3Enter the output tokens you expect and how many identical runs you plan.
- 4Click Estimate cost and read the breakdown, shown to 4 decimal places.
Model: GPT-4o (snapshot price $2.50 per 1M input tokens, $10.00 per 1M output tokens). Input: 1,000 tokens, so input cost = 1,000 / 1,000,000 x $2.50 = $0.0025. Output: 500 tokens, so output cost = 500 / 1,000,000 x $10.00 = $0.0050. Total for one run: $0.0075.
- Prices are a snapshot dated November 2026, labeled on the tool: "Snapshot: November 2026. Prices and limits change; verify with the provider before budgeting."
- The open-weights Llama entry is priced at $0 because you run it yourself; your real cost is your own hardware or hosting, which this calculator does not estimate.
- Real bills can differ: providers may add charges for caching, tools, images or minimums, and output length is always a guess until the call is made.
They were recorded in the November 2026 snapshot and are labeled as such. Providers change prices, so check the provider's pricing page before committing a budget.
About this tool
The Prompt Cost Calculator estimates what a prompt would cost on a pay-per-token API, across a small table of well-known models. Give it input and output token counts (or paste your prompt and let it count the input), pick a model, and it shows the input cost, output cost and total, worked out from the snapshot price table. It is an estimating tool for planning, never a live quote.
Related tools
Part of Ovatly's free tools in AI & Prompt Tools.