Calculators

Calculators for AI agents and automation

Plan the cost, reliability and return of AI agents and automation with your own numbers. Every formula is shown, and the math runs in your browser.

Cost and tokens

Estimate LLM API cost per request, day and month for Claude Opus 5.5, GPT-6.1 Sol, Gemini 4 Argon and other models, with current prices per 1M tokens

See how steps and a growing context drive the token count and cost of one agent run and a month of runs

Compare monthly input cost with and without prompt caching, using your cache discount, write premium and hit rate

Turn pages and words into an estimated token count and see how much of a model’s context window your documents fill

Plan a retry policy: see each wait, the longest delay and the total time a step can spend retrying before it gives up

See how a small failure rate per step compounds over a long agent run, and how retries change the odds

Size the workers behind a job queue from arrival rate, processing time and the utilization you want to keep

Find out how many agent runs and parallel agents your API rate limits allow, and whether requests or tokens are the bottleneck

Turn an uptime percentage like 99.9% into the downtime it allows per day, week, month and year