Cursor doubled included model usage on all plans, adding Grok 4.5 and Composer 2.5. The change came via a social media post without pricing details.
Key facts
- Doubled included usage of Cursor models on all plans.
- Added Grok 4.5 and Composer 2.5 models.
- Announced via X, no blog post or changelog.
- No pricing changes disclosed.
- GitHub Copilot charges $10–$39/month for similar access.
Cursor, the AI-powered code editor, announced a usage increase and model expansion on all subscription tiers. According to @cursor_ai's X post, the company doubled the included usage of Cursor models across every plan. The update adds Grok 4.5 and Composer 2.5 to the available model roster, expanding the options developers can leverage without extra charges.
No pricing changes were disclosed alongside the usage increase, suggesting Cursor is using model access as a competitive differentiator rather than a direct revenue lever. The announcement came via a repost by @cursor_ai of @leerob's original post, with no accompanying blog post or changelog entry — an unusually terse communication for a meaningful product change.
Cursor competes with GitHub Copilot, which offers tiered model access at $10–$39 per month but has not recently announced similar usage increases. Doubling usage without raising prices pressures competitors to match or risk losing cost-sensitive developers, especially as AI code generation becomes table stakes for modern IDEs.
What the move signals
The addition of Grok 4.5 — xAI's latest model — is notable. Grok has been positioned primarily as a consumer chatbot, not a code-completion engine. Cursor's inclusion suggests either a custom fine-tune or a broader partnership that could give xAI a foothold in developer tools. Composer 2.5, meanwhile, is Cursor's own model lineage, indicating continued investment in proprietary models over reliance on third-party APIs.
Developers on the free tier will benefit most: doubled usage with no price hike lowers the barrier for trying Cursor's models versus GitHub Copilot's limited free tier. However, Cursor did not specify the new usage caps in absolute terms (e.g., requests per month), making direct comparison impossible without user testing.
Broader context
This move follows a pattern of AI coding tools competing on usage limits rather than raw capability. In the past 90 days, GitHub Copilot expanded its multi-model support, and Replit introduced its own model-agnostic agent. Cursor's doubling of usage without a price increase is a direct bid to retain users who might churn to cheaper or more generous offerings.
For xAI, the Grok 4.5 integration is a low-risk test of enterprise appeal. If developers adopt it for code generation, xAI could push for deeper integration, including potential training on code data — a valuable but contested dataset.
What to watch
Watch for Cursor to publish updated pricing pages or changelogs with absolute usage caps. Also watch for GitHub Copilot to respond with a similar usage increase or model addition, particularly if Cursor sees a measurable user retention lift in the next 30 days.
