Cursor (X Subscription)
Coding plans · reviewed by Stan K. · Source: data snapshot · data as of 2026-11-11 · scored at Premium
The IDE plan with the cleverest trap: the weekly pool covers only Grok and Composer — Claude, GPT and Gemini ride your prepaid credit at API list prices.
#10 of 10 in AI Coding Plans
Tiers — what each dollar buys
| Tier | $/mo | Annual | Quota & notes |
|---|---|---|---|
| Free | $0 | — | Hobby tier; no third-party models |
| Plus | $8 | — | Grok models + Composer in the weekly pool; NO third-party models (Claude/GPT/Gemini excluded entirely) · annual $80 |
| Premium scored | $30 | — | Weekly pool 15× Plus + third-party models at API list prices from prepaid credit · annual $300 |
| Super | $100 | — | 50× Plus pool · annual $1,000 |
| Ultra | $200 | — | Top tier · annual $2,000 |
Agent support: Cursor IDE only — The plan fuels Cursor's own IDE/agent. No third-party CLI support from the subscription.
PlanScore breakdown — why 53/100
10 plan criteria × 10 points; PlanScore = mean × 10. Low pillars are visible on purpose — that's the point of a score you can audit. Methodology →
What it does well
- ▸Polished IDE + agent with a huge user base — the workflow many developers already live in
- ▸Grok 4.5/4.6/4.7 and Composer 2.5 included in the weekly pool at every paid tier from $8
- ▸Annual billing cuts the price ~17% ($80/$300/$1,000/$2,000)
- ▸Third-party frontier models (Claude, GPT, Gemini) available on Premium+ — just not in the pool
Where it costs you
- ▸The weekly pool excludes exactly the models people pay frontier prices for — Claude/GPT/Gemini are metered at API list from prepaid credit
- ▸Plus and Free have no third-party models at all
- ▸Pool size published only as relative multipliers (4×/15×/50×) — opaque in absolute terms
- ▸October 2026 pricing rework — the second major pricing change in a year
Hidden traps — the stuff the pricing page doesn't say
The subscription's included usage covers Grok 4.5/4.6/4.7 and Composer 2.5 only. Use Claude, GPT or Gemini and you pay API list prices out of a prepaid balance — the "all-you-can-eat" feeling applies to the models Cursor wants you to use.
Cursor publishes 4×/15×/50× relative tiers but not the absolute pool size — you cannot compute what you're buying before you buy it.
The pool refreshes weekly, not every 5 hours — a two-day intensive can leave you five days on prepaid credit.
The October 2026 'X Subscription' rework is the second major repricing within a year — terms have moved under subscribers before.
Know the walls before you pay: read the traps above, then decide with eyes open.
See Cursor (X Subscription) plansAffiliate link where available — never affects the score.
- official Cursor models & pricing docs (official) verified
- official Cursor pricing page (official) verified
- community HN/community: Grok force-out + generosity complaints (via recon, 2026-11-11) unverified
Data as of 2026-11-11. Prices and limits change monthly — report anything stale. How we verify: methodology.
Closest alternatives
Gemini 4 in Antigravity for $19.99 — until the 10-day lockouts, the opaque agent tiers and the Code
Codex rides the ChatGPT plans — and since the limits merged, your chat, Work and Excel share one opa
~1.7B tokens/month for the $20-equivalent tier — but you can only buy it a year at a time, and the q
Cursor (X Subscription) plan review
In-depth analysis · last updated 2026-11-11
Tiers and discounts
Free $0 / Plus $8 / Premium $30 (scored) / Super $100 / Ultra $200, annual billing at $80/$300/$1,000/$2,000 (≈ −17%). The October 2026 'X Subscription' rework moved everything to a weekly pool + prepaid credit model.
The trap, plainly
The weekly pool includes Grok 4.5/4.6/4.7 and Composer 2.5. It does not include Claude, GPT or Gemini — those are available from Premium up and metered at API list prices from your prepaid balance. Plus and Free tiers have no third-party models at all. So the advertised price is real for the models Cursor prefers, and something else entirely for the frontier models the marketing shows. We score transparency 3 — the pool's absolute size is published only as 4×/15×/50× multipliers.
Volume and community
No verified token counts exist (the pool is opaque by design). Community sentiment clusters on competitors feeling more generous and frustration at being pushed toward Grok — labeled sentiment, not a measured range.
Worst value tier
Plus $8 — it looks like the value tier but excludes third-party models entirely; you're buying a Grok/Composer-only subscription inside an IDE.
Versus OpenCode Go and Copilot
OpenCode Go ($10) gives a dollar-denominated budget across 32 models in open CLIs; Copilot Pro ($10) gives $15 credits across Claude/GPT/Gemini/Grok/Kimi with CLI access. Cursor Premium ($30) gives a Grok-shaped pool plus list-price metering for everything else — the weakest value story of the three at comparable prices, unless the Cursor IDE workflow itself is what you're buying.