Auto mode in Cursor is now powered by Cursor Router, an intelligent model routing system that analyzes each request and directs it to the appropriate model based on task complexity. Users can select from three optimization modes—Intelligence, Balance, and Cost—each trading off quality and price differently along a cost-intelligence frontier. Admins gain controls to enable the router per team or group, restrict optimization modes, set defaults, and allow or block specific underlying models. The router is available across desktop, web, iOS, CLI, and the SDK, enabled by default on Teams plans, with Grok 4.5 required as one of the price-efficient routing options.
Questions this post answers
What is Cursor Router and how does Auto mode work now in Cursor?
Cursor Router is an intelligent model routing system that now powers Auto mode, analyzing each request and sending it to the right model based on task complexity. Frontier models handle demanding work while price-efficient models handle simpler tasks. Users choose an optimization mode—Intelligence, Balance, or Cost—to control where requests land on the cost-intelligence tradeoff, and Balance and Intelligence modes bill at the routed model's rate. Teams weighing model choice and cost in their coding workflow can follow router changes like this on daily.dev.
What optimization modes are available with Cursor Router and how do they differ?
Cursor Router offers three optimization modes: Intelligence, which matches the most expensive and powerful frontier models; Balance, which matches frontier models most people use daily; and Cost, which reaches the highest available intelligence while optimizing token spend. Balance and Intelligence bill at the routed model's rate, and each mode positions requests differently along the cost-intelligence pareto frontier. Developers comparing AI coding tool pricing and quality tradeoffs can track updates like this on daily.dev.
Can admins control which AI models Cursor Router uses for a team?
Yes, admins can enable Cursor Router per team or group, restrict which optimization modes members can use, set a default mode, and allow or block specific underlying models. The router is on by default for Teams plans, while Enterprise admins must enable it from the dashboard, and it supports both soft and hard enforcement for standardizing on Auto mode. Engineering leads rolling out AI coding tools org-wide can follow admin control changes like this on daily.dev.