Cursor announced Cursor Router, an automated model-routing feature that the company claims can choose the “right model for the task at hand” and deliver “frontier-quality results at 60% lower cost.”
In Cursor’s description, users can pick Auto mode and then optimize for “Intelligence,” “Balance,” or “Cost.” The company says Router sends “frontier models when the work demands them” and “price-efficient models when it doesn't.” In early access, Cursor also claims customers saw “no drop-off in quality” alongside a lower cost per commit compared with routing all requests to Opus 4.8. One chart accompanying the launch is titled “Cursor Router produces frontier-quality results at a 60% reduction in cost,” with a subtitle claiming a “20% Satisfaction improvement vs. Opus 4.8.” The same material maps “Balance” to “36% Savings” and “Intelligence” to “60% Savings,” relative to Opus 4.8.
Cursor says Router is available now across all surfaces on Teams and Enterprise plans. Admins can enable it per team, allow or block specific models, set defaults, and disable select optimization modes for users. A separate settings panel also includes an “Underlying model” option that can display which model Auto routed to at the start of each response, though Cursor recommends keeping it hidden “to avoid model bias.” That setting is listed as available only for Balance and Intelligence modes.
Source: Cursor
