Cursor's new plans bundle Grok and bill Claude and GPT at API list price: what teams should check

From 12 October, new Cursor customers join The X Subscription, a Grok-first bundle where Claude and GPT models are billed per use at list price. Here is what the Cursor pricing change means for teams budgeting for AI coding tools.

Cursor changed how new customers pay on 12 October. New sign-ups now go onto The X Subscription, a bundle of Grok, Cursor and X Premium. Existing subscribers are not moved automatically, and switching is permanent once done.

The published tiers run from a free plan through Plus at $8 a month (introductory), Premium at $30, Super at $100 and Ultra at $200, with Teams starting at $30 per user per month. The included usage is for Grok models. Claude and GPT are no longer part of the allowance. On Premium and above you can still select them, but they bill from the first request against prepaid credit, at the model makers’ list prices. One analysis cites Sonnet 5.5 at $2 input and $10 output per million tokens, and Opus 5.5 at $4 and $20. Cursor has not explained how prompt caching is billed, so treat any cost maths as an estimate.

What this tells you

Until recently, a flat monthly seat felt like a flat monthly cost. This is a move toward metered pricing for the best-known models, with a cheaper in-house-adjacent model as the default. It follows the same direction as the other AI coding subscription changes we have tracked this month: the headline price stays low, and the real bill depends on which model your developers pick and how hard they run agents.

Two practical effects for anyone commissioning software:

So what

If you pay an agency or contractor who uses Cursor, ask two questions: which models do they use for which work, and are AI tool costs inside the fixed price or passed through? A good partner can answer both in a sentence, and will have a budget view of agent usage rather than a surprise at month end. Our AI-assisted development approach prices the work as a whole, so tool-pricing changes like this are our problem, not yours. If you want a second opinion on a quote, start a project conversation.

← All research

More research.

GitHub's Copilot metrics now count OpenCode and Claude Code traffic: who really uses which AI coding tool

On 23 October 2026 GitHub added a per-tool breakdown to the Copilot usage metrics API, so tools such as OpenCode and Claude Code are counted even without editor telemetry. It shows how mixed the AI coding tool stack inside software teams has become.

Read ›

GitHub Copilot retires Claude Opus 5 and GPT-5.6 Terra on 16 November: the model-churn bill for buyers

GitHub has announced that Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra leave every Copilot experience on 16 November 2026. For anyone commissioning AI-assisted development, model retirement is now a monthly event to plan around.

Read ›

GitHub Copilot now benchmarks how 'agent-first' your developers are against your industry

On 22 October 2026 GitHub added opt-in industry comparisons to its Copilot impact dashboard, ranking firms by share of agent-first developers and by the merge rate of agentic pull requests. Here is how to read the numbers when you are buying software.

Read ›

Apple's new bait-and-switch rule 2.5.19: what it means for AI-built and fast-shipped apps

On 16 October 2026 Apple updated its App Review Guidelines: apps that change after approval can be removed from the App Store and the Developer Program, and AI-generated intimate imagery is now explicitly covered. Here is what founders commissioning an iOS app should check.

Read ›

GitHub Copilot agentic autofix is generally available: AI now fixes security alerts, and bills for it

GitHub made agentic autofix for code scanning alerts generally available on 15 October 2026. Copilot can now fix up to 25 security alerts in one pull request, using AI credits your supplier or your team pays for.

Read ›

GitHub AI Scan for pull requests goes GA: AI security checks for PHP, Terraform and Dockerfiles

GitHub's AI Scan for pull requests became generally available on 15 October 2026, checking languages the standard scanner misses. Findings are advisory and cannot block a merge, which matters when you buy software.

Read ›

Start a project

Building with AI? Let's talk.

Tell us what you are building. We'll give you an honest view of the right approach.

Start a projectSee our work