Tools · AI cost
What will your
AI agent actually cost?
Model the monthly API spend of any LLM workload across Claude, GPT, and Gemini — with
prompt-caching savings factored in. No signup, runs entirely in your browser.
Same workload, every model
Your inputs above, priced across each model. Cheapest first.
| Model | In $/1M | Out $/1M | Per request | Per month |
Prices as of 2026-06-29, per million tokens. Confirm against each provider's live pricing
before budgeting.
How to read this
Most agent budgets blow up in one of two places: sending a giant prompt on every request, or
running a frontier model on steps a cheaper one handles fine. This calculator makes both
visible. Drop your real per-request token counts in, then watch the comparison table — the gap
between tiers is usually larger than people expect, and prompt caching often moves the needle
more than switching models.
For the full reasoning behind these trade-offs, see
AI agent cost math: when Haiku beats Sonnet
and
prompt caching without switching models.
Want it built right?
I build cost-tuned agent systems for founders — the cheapest model that does the job,
caching wired in from day one.