Private Everything runs in your browser. No uploads, no tracking, no signup./ zero data collected

Tools / 02 · AI Cost Calculator

AI Cost Calculator

Pick a model, enter tokens per request and monthly volume, and see the bill before you build.

Model System / context tokens per request (sent every call) New input tokens per request (this turn) Output tokens / request Requests / month Use prompt caching (cached tokens billed at ~10%)

—

Context tokens (system prompt + conversation history + reused code) are resent on every call, so they dominate spend. Caching the stable prefix cuts that portion ~90%. Prices are indicative per-million-token rates (USD) and change often.

Compare the best vibe coding tools →

Cut your AI coding bill with prompt caching →

Frequently asked questions

Are the prices accurate?: They are indicative per-million-token rates captured at build time. Providers change pricing often and use tiered or cached pricing, so treat the result as a planning estimate and confirm on the official pricing page before committing a budget.
Does this calculator upload my inputs?: No. All math runs in your browser. No token counts, model choices, or volumes leave your device.
Which models are included?: The most common API models from OpenAI, Anthropic, Google, DeepSeek, and Meta (hosted). It is not exhaustive — add your own mental math for anything missing.
How do I lower my bill?: Use a smaller or mini model for cheap, high-volume tasks, cache long system prompts, and keep outputs short. The calculator shows exactly how much each lever saves.