Generation allowance
The generation allowance is the number of model calls your plan includes per day and per month. Reach for this page when a prompt is refused, or when you want to know what a turn costs before you send it.
Quick start
usageType that in the terminal and it prints the month and the day, used and limit, with both reset times in UTC. Settings shows the same line under Generation allowance.
How it works
What spends a unit
Every model call spends one unit, in the terminal and in chat mode alike. A terminal turn can make many calls, and each one counts. The 17 workspace commands such as build and preview-open, typed by you, run with no model call and spend nothing.
The two ceilings
The allowance has a daily ceiling and a monthly one, and a prompt is refused when either is spent. The daily ceiling resets at the next UTC midnight and the monthly one on the first of the next UTC month. The numbers below are the plan defaults.
What keeps working
When the allowance is spent, only generation locks. The extension keeps running in the emulator, and editing in the Code tab, checkpoints, download and share stay open. The refusal card says so, and offers Download your extension.
The refusal card
The card is titled Today's generation allowance is used up or This month's generation allowance is used up. It says when the ceiling resets, reads Paid plans start on the first charge date with a link to the pricing page, and its main button is Notify me.
The generation size ceiling
A second ceiling bounds the size of what the terminal generates per UTC day: 2,000,000 tokens. A call over it is refused as TOKEN_CEILING without spending a unit, and the same card appears with the title Today's generation size limit is reached. It resets at the next UTC midnight.
The turn cap
One terminal turn can make at most 40 model calls. The 41st is refused as TURN_CALL_CAP with no unit charged, and the message says to send a new prompt to keep going.
Reference
| Plan | Per UTC day | Per UTC month |
|---|---|---|
| Free | 50 | 300 |
| Pro | 200 | 2000 |
| Limit | Value |
|---|---|
| Units per model call | 1 |
| Model calls per terminal turn | 40 |
| Tokens generated per UTC day | 2,000,000 |
| Refusal | Meaning |
|---|---|
QUOTA_EXCEEDED | The day's or the month's allowance is spent |
TOKEN_CEILING | The day's generation size is spent, no unit charged |
TURN_CALL_CAP | This turn made 40 calls, no unit charged |
When it fails
- The card says today's allowance is used up. Wait for the next UTC midnight, or move to Pro. What you built stays on screen, and download and share still work.
- The card says this month's allowance is used up. The monthly ceiling resets on the first of the next UTC month.
- The card says today's generation size limit is reached. The terminal generated 2,000,000 tokens today. No unit was charged. It resets at the next UTC midnight.
- A turn stops after many calls. The turn hit 40 model calls. Send a new prompt and the agent continues.
usagesays to sign in. Your session ended. Sign in and run it again.
Next
- Terminal for the commands that spend nothing
- Chat mode for what one prompt spends
- Settings and sign-in for the Generation allowance row