AI Prompt Cost Calculator
What does this prompt cost at 10K, 100K, 10M calls a day — on Claude, GPT, Gemini or open models?
Pick a model, set input and output tokens per call, set a cache-hit rate, and get the per-call, per-day, per-month and per-year cost — across all major providers for comparison.
Cached input is ~10% of normal cost on providers that support it.
Runs entirely in your browser — nothing you enter is sent to a server.
Need this for real, on your stack?
These free tools are a taste of how we think. We’re a senior software team across Romania & Pakistan that ships deep technical work — platforms, infra, data and the gnarly bits in between.
Talk to our engineers →Cache-hit rate is the lever
For repetitive workloads (system prompts, RAG context), prompt caching cuts input cost to ~10% of base. A 70% cache hit rate cuts your bill by ~60% with zero quality change.
Rates
Prices are a 2026 snapshot stored in weww-tools.js — edit there to reflect current quotes.
Built by WeWorkWorldwide — a senior software team in Romania & Pakistan that ships deep technical work. Need this turned into something production-grade for your stack? Talk to our engineers →