Everything you need to know about using AIModelCalc
AIModelCalc is a free, browser-based tool that lets you calculate and compare AI API costs across all major providers — OpenAI, Anthropic, Google, Meta, Mistral, xAI, and Cohere. No signup required, no data sent to our servers.
Yes, completely free. AIModelCalc is supported by display advertising. You'll never be asked to create an account or pay anything.
No. All calculations run entirely in your browser using JavaScript. Your token counts, request volumes, and any other inputs are never transmitted to our servers and are gone when you close the tab.
We aim to keep pricing current with official provider documentation. However, AI providers change their prices frequently — always verify against the official pricing pages before making financial decisions. Our data is updated regularly and dated in the hero badge.
We currently track 14 models across 6 providers: OpenAI (GPT-4o, GPT-4o mini, GPT-3.5 Turbo, o1, o3-mini), Anthropic (Claude 3.5 Sonnet, Claude 3.5 Haiku, Claude 3 Haiku, Claude 3 Opus), Google (Gemini 1.5 Flash, Gemini 1.5 Pro, Gemini 2.0 Flash), Meta (Llama 3.1 70B, Llama 3.1 8B), Mistral (Mistral Large, Mistral 7B), xAI (Grok-2), and Cohere (Command R+).
We review and update pricing at least monthly, and sooner when major providers announce changes. The last update date is shown in the "Updated — Live pricing" badge at the top of the homepage.
Yes. The Cost Calculator (Tool 1) includes a "Cached Input %" field. Cached tokens are priced at a 90% discount from the standard input rate, consistent with how OpenAI and Anthropic handle prompt caching.
Tool 1 lets you enter your expected token usage — input tokens, output tokens, monthly request volume, and cached input percentage — and instantly calculates your cost per request, monthly total, and annual total for any model.
Tool 2 shows all models side-by-side with their input price, output price, context window size, speed rating, and best use case. You can filter by provider and sort any column to find the right model for your needs.
Tool 3 lets you select a use case (Customer Chatbot, Code Copilot, Doc Summarizer, etc.) or enter custom parameters, then shows you what you'd pay per month across all supported models — ranked from cheapest to most expensive.
The context window is the maximum number of tokens a model can process in a single request (prompt + response combined). Larger context windows let you send longer documents or maintain longer conversations without truncation.
The speed dots (● ● ● ◌ ◌) represent a relative speed rating for each model based on typical tokens-per-second throughput. More filled dots = faster output. This is a qualitative indicator, not a guaranteed benchmark.
Premium models like Claude 3 Opus and o1 are designed for complex reasoning tasks and are priced accordingly. Smaller models like Gemini 1.5 Flash and GPT-4o mini are optimized for speed and cost at high volume. The best model depends on your quality requirements and budget.
Some providers offer enterprise pricing or volume discounts for high-usage customers. The prices shown in AIModelCalc reflect standard pay-as-you-go API rates. Contact individual providers directly for enterprise pricing.
Please email us at hello@aimodelcalc.com with the model name and the correct price from the provider's official documentation. We'll review and update as quickly as possible.