Skip to content
Multilo Docs
Changelogmultilo.com
Download

Costs and usage

Track how many credits you've used and on what, cap your spending with extra usage charges and a spend limit, and get more done with fewer credits.

Multilo meters AI work in credits. This page shows where to see how many you’ve used and on what, how to cap your spending, and the habits that get more done with fewer credits. For plan prices and allowances, see the pricing page.

How AI work is metered

  • 1 credit = $0.01. Your plan includes a monthly allowance: 25 credits on Free, 1,000 on Pro, and 3,900 on Pro+.
  • You pay for the work a request takes. Each request to a model is priced by the tokens it reads and the tokens it writes, at that model’s rate. A quick edit costs very little; a long agent run, which makes many requests, costs more.
  • Repeated context is cheaper. When the same context is reused across requests in a session, the repeated part is billed at a cheaper cached rate.
  • Local work is free. Opening and editing files, viewing images, browsing your library, and searching by keyword run on your computer and use no credits.

Track your usage

In the app

  • Your balance is shown in the app. While it’s confirming the latest figure, it shows a brief placeholder rather than a stale number.
  • The AI Credits panel breaks down your usage for the current period by feature and by model.
  • /models in chat lists every model Auto used in the session.

On multilo.com

Open Account → Usage:

  • AI credits this period — how many you’ve used (and what that’s worth in dollars) and how many remain. Any bonus credits are shown too; they don’t expire at the monthly reset.
  • Where your credits went — your credit use this period by feature, with the number of calls, the credits, and the model for each. Once enough of the period has passed, it also projects how many credits you’ll use by the end of it.
  • Download CSV — export the breakdown for your own records.

Control your spending

When credits run out

Included credits reset at the start of each billing period, and unused credits don’t roll over. If you run out before then, AI requests pause until the next period — unless you’ve turned on extra usage charges. Nothing is charged without your choice. On Auto, Multilo shifts to a lighter model as your balance runs low, so you can keep working for longer.

Extra usage charges and spend limit

On Pro and Pro+, you can keep using AI after your included credits run out. Set it up in Account → Subscription:

  1. Add a payment method

    Extra usage needs a card on file.

  2. Turn on Extra usage charges

    Extra credits are billed at $0.01 each, at the end of the period. Turn it off to stop at your included credits instead.

  3. Set a monthly spend limit

    Choose the most you’ll be billed for extra usage each month. The page shows how much of it you’ve spent this period.

The spend limit is a hard stop

When you reach your monthly spend limit, AI requests stop until the next period — it isn’t just a warning. Without a limit, you’re billed for all extra usage. The Free plan has no extra usage: it stops at its included credits.

Per-task token budget

To keep one long agent run from spending more than you expect, set a per-task token budget in Settings → AI in the app. It’s off unless you set one. When a run reaches it, the agent pauses and asks whether to continue.

Reduce credit use

The cost of a request grows with what the model reads and writes. These habits keep it down:

  • Match the model to the task. A fast model is fine for tidy-ups and summaries; save flagship models for critique and long drafts — or let Auto decide. See choosing a model.
  • Pick the right MODE. A light MODE for a quick pass; a deep one only when the task needs it.
  • Start a new chat for a new task. A long chat sends its whole history with every request.
  • Point, don’t paste. @-mention the one source that matters, or scope the chat to a folder, instead of pasting whole papers.
  • Keep standing instructions short. Everything in AGENTS.md is read with every request. Move specialized procedures into skills, which load only when needed.
  • Plan big changes first. A plan you approve is cheaper than redoing a chapter that went the wrong way.
  • Stop early. If a run is heading the wrong way, press Stop or Esc rather than letting it finish.
  • Be specific. “Tighten section 2.3 to 400 words” costs less than “improve chapter 2”.

Why usage climbs in a long chat

  • Long context: every request carries the conversation so far. The repeated part is cheaper, but a one-line question late in a long chat still reads the whole history.
  • Agent runs: an agent makes many model requests — read, reason, draft, revise — and each one is metered.
  • Big inputs and outputs: whole documents, many sources, and long drafted sections all add tokens.
  • Deeper work: flagship models and deep MODEs cost more per request.

Check Where your credits went to see which features and models used the most, and adjust.

Billing questions

CreditsWhat a credit is and what uses one.Model configurationChoose and switch models.Auto model selectionThe right model for each message.Best practicesKeep context focused and work in steps.