Grok 4.7
Grok 4.7 is a first-party Cursor model, built together with SpaceXAI for coding and other knowledge work. It builds on Grok 4.6 with a larger base, longer training on harder multi-hour tasks, and stronger self-verification.
What is Grok 4.7?
Grok 4.7 is designed for difficult, long-running tasks that require creative tool use, whether in software engineering, data science, finance, legal work, or anything else you do on a computer. It stays with complex multi-step work and checks its own results more carefully than Grok 4.6.
- General model for coding and general knowledge work
- Larger base than Grok 4.6, trained on harder multi-hour tasks
- 256k standard context window, with 500k long context and long-context rates above 256k input tokens
- Stronger self-verification on long-running agent work
- Four effort levels: low, medium, high, and xhigh
While Composer remains available for everyday coding speed and cost, Grok 4.7 is the general model aimed at harder, longer sessions.
What are Grok 4.7 effort levels?
Grok 4.7 supports four effort levels: low, medium, high, and xhigh, so you can match compute to how hard the task is. Higher effort lets the model think longer on difficult tasks. Effort levels are more separated than in Grok 4.6. The named-model default is high. When using Auto, Cursor uses the same default (high) at launch.
On Cursor Start, Grok 4.7 is fixed at medium effort in non-fast mode. You cannot change the effort level or enable Fast mode on that plan. Upgrade to Pro or higher to choose effort levels and Fast mode.
When should I choose Grok 4.7 over Composer?
- Choose Grok 4.7 for harder, longer-running work: multi-step agents, large codebases, and tasks that stalled or felt short on Grok 4.6.
- Choose Composer for everyday coding where speed and cost matter most. Composer is not being deprecated.
Which plans include Grok 4.7?
Grok 4.7 uses the same first-party credit treatment as Grok 4.6. Credits that apply to Grok 4.6 also apply to Grok 4.7. Included plan usage from the first-party pool covers 4.7 the same way.
Every paid plan includes access to Grok 4.7: Start (India), Pro, Pro+, Ultra, Teams, and Enterprise. Grok 4.7 draws from the Cursor Models usage pool included with your plan. On Start, Grok 4.7 runs at a fixed medium effort level in non-fast mode.
Where can I use Grok 4.7?
Grok 4.7 is available everywhere you use Cursor:
- Desktop app
- Web and cloud agents
- CLI
- SDK
- Automations
- iOS app
Which regions support Grok 4.7?
Grok 4.7 is available in every country where Cursor normally offers models, including the EU. See Regions and model availability for what to do if a model is unavailable in your region.
Is Composer being deprecated?
No. Composer remains available for everyday coding speed and cost; Grok 4.7 is the general model for harder, longer work.
Why can't I see Grok 4.7?
Enterprise team admins manage model access at cursor.com/dashboard/team-settings/models. See model and integration management for details.
If Cursor models or Grok / SpaceXAI models are blocked for your team, an admin must enable them before members can use Grok 4.7.
Grok 4.7 is served under the Cursor vendor catalog, so teams that previously blocked the xAI provider can still enable 4.7.