The week: 1 hour of agentic coding and 10 quick questions every weekday.
ForecastX Subscription PremiumLow confidence
Fog: no public number to forecast against.
We can't forecast this from public data. X Subscription (Grok + Cursor) publishes only relative limits (4x Plus) for X Subscription Premium. Calibrate with one reading from your own meter.
- Mon (Monday) FogNo public cap to compare $8.20$2.10–$25.50
- Tue (Tuesday) FogNo public cap to compare $8.20$2.10–$25.50
- Wed (Wednesday) FogNo public cap to compare $8.20$2.10–$25.50
- Thu (Thursday) FogNo public cap to compare $8.20$2.10–$25.50
- Fri (Friday) FogNo public cap to compare $8.20$2.10–$25.50
- Sat (Saturday) FogNo public cap to compare $0no API-$ work
- Sun (Sunday) FogNo public cap to compare $0no API-$ work
What is published
4x the Plus pool, shared by Grok Chat, Imagine, Voice, Grok Bot and Cursor; third-party models in Cursor billed at API list price.
Multiplier: 4x Plus. Window: weekly shared pool. Multipliers only (4x, 15x, 50x Plus). The size of the base weekly pool is not published.
How to calibrate X Subscription Premium
- Open the calculator with this week loaded and X Subscription Premium selected.
- Work one normal day and note how much of your weekly meter it used.
- Enter “Day [that day] of my plan used [__]% of my weekly meter”. The calculator turns your reading into an implied weekly capacity in API-$ and forecasts the rest of the week.
- Remember what shares the pool: 4x the Plus pool, shared by Grok Chat, Imagine, Voice, Grok Bot and Cursor; third-party models in Cursor billed at API list price. Image and video work is counted in vendor credits and not modelled here.
Caveats and confidence
Low confidence Multipliers only (4x, 15x, 50x Plus). The size of the base weekly pool is not published.
Figures used: official. See the plan page for each figure's basis and source.
- Task costs are ranges. The widest is “Agentic coding, 1 hour, one interactive agent”, at ±5-10x (effort level and number of parallel agents), and it dominates most coding weeks.
- Vendors change limits often. See the changelog; all figures were checked 20 Oct 2026.
- Read the methodology for how the hit-day range, off-days and the session check work.
FAQ
Is X Premium (Grok + Cursor) enough for daily coding?
We can't compute this from public data. X Subscription (Grok + Cursor) publishes only a multiplier for X Subscription Premium (4x Plus), not the size of the pool, so below is how to calibrate it with one reading from your own meter.
Which day would X Subscription Premium run out on this week?
Fog: no public number to forecast against. We can't forecast this from public data. X Subscription (Grok + Cursor) publishes only relative limits (4x Plus) for X Subscription Premium. Calibrate with one reading from your own meter.
What happens when you hit the limit on X Subscription Premium?
Top-ups at per-model rates. The overflow price is not published.
What does X Subscription (Grok + Cursor) officially say about X Subscription Premium limits?
4x the Plus pool, shared by Grok Chat, Imagine, Voice, Grok Bot and Cursor; third-party models in Cursor billed at API list price.
How reliable is this answer?
Low confidence. Multipliers only (4x, 15x, 50x Plus). The size of the base weekly pool is not published.