Guides
See what a Muse Code thread cost
Open the usage page in Helicon. It shows what each thread would have cost at Meta's published per token rates, grouped by day, by model and by thread, alongside the 5 hour window and weekly cap as Muse Code reports them. It is a meter reading, not a bill.
Last updated · Markdown version
What you are looking at
Your subscription already paid. These figures multiply the tokens each turn used by the published rate for the model that used them, which tells you which work is expensive. Nobody sends you this number as an invoice.
Reading the three breakdowns
| View | Question it answers |
|---|---|
| By thread | Which piece of work was expensive |
| By day | Whether today was unusual |
| By model | What reasoning effort actually costs |
Using it to change behaviour
- Compare a high effort run against a low effort one on similar work.
- Look at threads where a subagent fan out ran, and see what the parallelism cost.
- Spot the long thread that has been re reading the same context every turn.
Frequently asked questions
Will Helicon ever bill me?
No. Helicon is free, MIT licensed, with no paid tier and no billing relationship.
Is this my plan usage?
No. Plan usage is shown separately as the 5 hour window and weekly cap, exactly as Muse Code reports them.
Is the calculation done locally?
Yes. Token counts come from Muse Code and the arithmetic happens on your machine.
Keep reading
- Cost at API ratesSee what every Muse Code thread would have cost at published per token rates, broken down by day, by model and by thread. A meter reading on your subscription, not a bill.
- Plan meterHelicon shows the Muse Code 5 hour window and weekly cap exactly as the CLI reports them, in the sidebar and on the usage page, so a limit never arrives as a surprise.
- PricingHelicon is free and MIT licensed with no paid tier, no account and no API key. It uses the Muse Code subscription you already pay Meta for.
Same Muse Code. Same subscription. Better interface.
Free and MIT licensed. Signed Windows installer, universal macOS DMG, and a web build against a daemon you run.