Grok subscribers say their weekly allowance now vanishes far faster than before. xAI has not confirmed any change, and its own documentation describes only the shared system introduced in June 2026.
A Reddit post on October 10 asked whether limits had dropped “by a lot” after one user exhausted a week’s quota in roughly half the usual time. Several other users reported similar experiences, according to the report.
What Users Are Reporting About the Weekly Quota
Several paying users describe one pattern: tasks that once cost a modest slice of the meter now cost about twice as much. One person said a single Grok Build prompt took several percentage points.
Earlier complaints surfaced in early October, when some SuperGrok subscribers said nearly half their allowance disappeared within a day or two. Every one of these accounts is anecdotal and unmeasured.
What xAI Has and Has Not Said
xAI has issued no statement addressing a reduction. Its Grok help page, last updated October 6, four days before the Reddit post, describes the weekly pool but mentions no recent adjustment to it.
The same page says a chat message “uses little compute” while video and long coding tasks use far more. That cost difference is documented, but how heavily each product is weighted is not published.
How the Weekly Usage Pool Works
Since June 2026, paid Grok users share one weekly pool across Chat, Imagine, Voice and Build instead of separate daily limits. Usage appears as a percentage, broken down by product, with a weekly reset time.
xAI, which its own pages now brand SpaceXAI, presented the change as simpler and more flexible. It said most people could go longer on a favourite product before hitting limits. It publishes no pool size for any plan.
Why Heavier Tasks Drain the Meter Faster
Because the pool measures compute rather than messages, one long coding session or high-quality video can consume more than dozens of text chats. Two subscribers doing different work can see very different burn rates.
That makes comparison hard. Someone who moved from chatting to Build, or from images to video, could watch the meter fall much faster this week even if nothing changed on xAI’s side.
The Weighting Question: Lower Limits or Different Costs?
A cut to the total allowance and a change in how features are weighted would look identical to a user watching the percentage bar. Only xAI can say which, if either, has happened.
Weighting changes would leave the headline percentage intact while making each action cost more of it. Reports of doubled costs fit either explanation, and nothing published so far separates the two.
What Is Documented and What Is Alleged
The table below separates what xAI’s documentation supports from what users claim. Keeping the two apart matters, because the allegation is the story, yet only the documented rows can be treated as established fact.
Readers should treat the alleged rows as unverified reports until xAI responds or independent testing produces repeatable numbers. No such testing has surfaced publicly, and no pool sizes exist to test against.
| Claim | Status |
|---|---|
| Shared weekly pool since June 2026 | Documented by xAI |
| Costs differ by product | Documented by xAI |
| Pool size per plan | Not published |
| Limits cut in October | Alleged, unconfirmed |
| Doubled cost per task | User-reported, unmeasured |
Expert Angle: Why Independent Verification Is Missing
No named analyst or independent lab has published measurements of Grok’s weekly meter. Without a fixed pool size, any test would need identical prompts run on identical plans before and after the alleged change.
READ MORE: Google Tests AI in Space With a Risky New Satellite
That is a high bar for ordinary subscribers. Until someone logs repeatable results with dates and screenshots, xAI’s documentation remains the only authoritative public description of how usage is counted, however incomplete.
What Happens When You Hit the Limit
Once the weekly limit is reached, paid features pause until the reset. Free-tier Chat and Voice limits stay available and run on their own separate schedule, according to xAI’s help page.
Subscribers can also buy Extra Usage Credits, from $5, or switch on Auto Top Up. xAI says those credits are priced at standard rates, so each action costs more than included usage.
How to Check Your Own Usage Meter
Anyone suspecting faster drain can open Settings, then Usage, on web or mobile. Recording the percentage before and after a fixed task gives a rough, personal measure of what that task costs.
Dated screenshots make those records useful to others. Repeatable entries are the only evidence that can move this dispute from anecdote toward proof, so log the task, the time and the starting percentage. The Usage tab shows:
- A progress bar with the weekly percentage used
- A breakdown by product: API, Build, Chat, Imagine and Voice
- The weekly reset date and time
- Your Extra Usage Credits balance, if any
Why Subscribers Are Frustrated
Paying users argue Grok’s weekly system has grown more restrictive without clear explanation. The complaints predate October and follow months of criticism over the move from daily limits to a single shared quota.
An opaque percentage bar offers little to dispute. Without published pool sizes, a subscriber cannot tell whether the plan now delivers less, only that the bar moves faster than before.
Implications for Paying Customers
If xAI did tighten costs, Grok subscribers on fixed monthly plans receive less for the same price, and heavier users face a nudge toward credits or higher tiers, both of which xAI’s help page describes.
If it did not, the episode still shows the risk of metering by hidden compute: users cannot audit changes themselves. Clearer published figures would settle disputes like this one quickly.
What to Watch Next
The decisive evidence would be an xAI statement, a changelog entry or a documented update to how products are weighted. Any of those would turn user suspicion into confirmed fact.
Independent, dated measurements would also help. Until either appears, the accurate description is simple: Grok users report faster drain on their weekly meter, and xAI has neither confirmed nor explained it.
