Tools · Token calculator

How much of a month of AI stays on your Mac?

Move the sliders. The first number is the tokens that stay on your own machine with chanio: the everyday questions. The second is what leaves it: the hard ones, kept and sent out once a night.

Questions per day30
Tokens in per question (question + the notes it reads)1,500
Tokens out per answer300
Share you’d keep for a bigger AI10%
YOUR MONTH · 900 QUESTIONS · 1,620,000 TOKENS
1,458,000
tokens stay on your Mac · 810 questions answered there
162,000
tokens leave your Mac · 90 kept questions, sent in one nightly batch

If every question went to an online AI, all 1,620,000 tokens would leave your Mac. Answers on your Mac use electricity and a one-time download of the small AI, and nothing is billed per token. The chanio starter sends only to Claude, with your own key; without one, it packs the kept questions for you to paste into any AI. Reasoning models also bill hidden thinking tokens, which this calculator does not count.

The meter is the product

This calculator is the same arithmetic chanio does for real, every day, in the meter under every answer: which tier ran and how many tokens. The point is that most questions never needed to leave your Mac, and you should be able to see which ones did. The founder’s own meter is public.

Questions people ask

How is it computed?
Questions per day × 30 days × (input tokens + output tokens). The share you keep for a bigger AI is the part that leaves your Mac; the rest is answered on your Mac and never leaves it.
What is a token?
Roughly three quarters of an English word. A typical question with the notes it needs to read is 1,000 to 3,000 tokens in; an answer is 100 to 500 tokens out. Reasoning models bill hidden thinking tokens as output, which this calculator does not count.
Is an answer on your Mac billed per token?
No: nothing is billed per token. You pay for the electricity, the one-time download of the small AI, and a Mac with enough memory. The small AI (a 3B model) on Apple silicon answers a short question in a few seconds.
Why keep anything for a bigger AI at all?
Because the small AI on your Mac is weaker at hard reasoning, current events, and long generation. chanio's honest answer is to keep those few questions for tonight, send them in one batch with your own Claude key or pack them for you to paste, and count what left, rather than pretend the small AI can do everything.

Related

Which AI model fits your Mac?
Pick a model by memory, with the honest trade-offs.
Set it up on your Mac
One line installs the starter and the meter.
Numbers, in the open
Last night's run and this site's visits, counted first-party.