Grok 4.7 in Cursor: Pricing and Uses

Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. is a frontier model from Cursor and SpaceXAI, aimed at coding and knowledge work. It builds on Grok 4.6A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product. Press Enter for the full definition. with a larger base, longer training on harder multi-hour tasks, and stronger self-verification. The table collects what Cursor publishes about it on the model page.

Three claims carry the pitch. It stays with difficult, long-running work, it checks its own results more carefully than Grok 4.6A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product. Press Enter for the full definition. does, and it manages longer context more reliably. The page does not say how the self-check works.

The two context numbers are easy to blur together. The 256k figure is the standard window, and the 500k figure is the long context, which is reachable but billed at higher rates once input passes 256k tokens. The pricing section below works through what that does to a bill.

Cursor's page publishes no benchmark figures for Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition., so this article repeats none. What the page does give is a direction from Grok 4.6A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product. Press Enter for the full definition.: more training on harder, longer tasks, on a larger base. Everything else below is plan, speed and price detail, plus some reasoning about how to use it.

Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. supports four effort levels: xhigh, high (the default), medium and low. Higher effort gives the model more time to work through a difficult task. Whether you get to choose depends on your plan, which the table sorts out.

Effort levels are more separated than they were in Grok 4.6A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product. Press Enter for the full definition., so harder tasks spend more time thinking. The page does not quantify the difference.

On Plus and on legacy Start there is no decision to make. Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. runs at medium effort at standard speed, so there is no toggle to hunt for. The Cursor StartAn India-only legacy individual plan at ₹649 a month centered on Cursor's own models, with cloud and mobile access but without Other Models, Auto, Bugbot, Automations or the Agent SDK; Cursor names Plus as the comparable plan on The X Subscription. Press Enter for the full definition. guide covers the India-only plan in more detail.

The same pricing page says Plus runs Cursor's own models at fixed settings: Fast is off, effort is medium, and Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. uses the 256k context window.

Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. draws from the weekly pool of Cursor models that every paid plan includes. Usage beyond the pool is priced per million tokens, and the table lays out all four combinations of speed and input length.

Long-context billing keys on input length alone. At or below 256k tokens of input, the full request uses the standard or Fast rates. Above 256k, the multiplier applies.

The consequence worth noticing is the cliff. A request with 250k tokens of input and one with 260k are nearly the same size, yet the second is billed at twice the standard rates. Long-context Standard ends up at the same per-token price as short-context Fast.

The step at the line is smaller for Fast. Fast input goes from $4 to $6, a 1.5x step, while Standard input goes from $2 to $4.

The page does not say how large the pool is on any plan.

The models and pricing page describes the pool more broadly: it is one pool of usage for Cursor models, shared across Agent in the editor, the CLI, Cloud AgentsAgents that run in a Cursor-managed virtual machine, check out the repo, do the work and open a pull request, then shut down, with no load on your laptop. Press Enter for the full definition., Grok BotAn early AI teammate product from SpaceXAI and Cursor that works across apps on a persistent cloud computer and asks for approval on selected actions; launch materials describe it as beta, but Cursor says Beta Services terms do not apply. Press Enter for the full definition. and the Grok apps, and it refreshes weekly. Remaining usage and the next reset date appear in your editor settings and on the usage page of grok.com/dashboard. When the pool runs out you can wait for the reset, keep working on prepaid credit at the same per-model rates, or move to a higher tier, and requests are never downgraded in quality or speed.

Cursor positions Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. for complex coding and multi-hour knowledge work, and says it has access to all agent tools when used with Cursor. The page names strengths rather than benchmarks.

The strengths Cursor lists are about duration and checking: staying with difficult, long-running work and checking its own results more carefully than Grok 4.6A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product. Press Enter for the full definition..

The tools line is short. The page says Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. has access to all agent tools in Cursor, then introduces a list that does not appear in the copy checked for this article. For what those tools are and how calls work, Cursor's agent overview and tool calling fundamentals are the places to look.

Cursor describes Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. as building on Grok 4.6A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product. Press Enter for the full definition., which is the comparison its page makes.

Our Grok 4.6A Cursor and SpaceXAI model for complex coding and knowledge work, still in the Cursor model pool and succeeded by Grok 4.7; it is a selectable model and is separate from the Grok Bot product. Press Enter for the full definition. guide lists the same $2 input and $6 output standard rates, so the two have the same standard-speed rates. One concrete difference is context: Grok 4.6 publishes a 256k window with no separate long-context tier, while Grok 4.7Cursor and SpaceXAI's current lead Cursor model, in the weekly pool at $2 per million input tokens and $6 per million output, with a 256k context window and a 500k variant; above 256k input tokens it bills at 2x the standard rate. Press Enter for the full definition. publishes a 500k long context. For the wider picture of picking between models, the Cursor model guide is the better starting point.