For two years, the model selector was the developer’s lever. You’d pick Sonnet for the hard stuff, something cheaper for the boilerplate, and the tradeoff was yours, request by request. Since July 22nd, Cursor took that lever without making a sound and handed a different one to whoever manages your team’s dashboard.
Auto mode now runs on Cursor Router. It classifies each request by task type and complexity, and sends it to the model that fits: frontier models for work that demands them, price-efficient models for everything else. That part is easy to accept. The part worth slowing down to look at is what “fits” means now, and who has the power to define it.
You pick Auto, and then you pick how the router optimizes. Three modes. Intelligence matches the most expensive frontier models, the ones you’d hesitate to use all day. Balance matches the frontier models most people use as their daily driver. Cost delivers good quality while reaching the highest intelligence possible without blowing your token budget. Each mode slides you over the same cost-intelligence Pareto frontier; all you choose is where you stop.
Here’s the billing detail that matters and the headline skips over: Balance and Intelligence bill at the rate of the routed model. So in those two modes, “let the router decide” doesn’t put a ceiling on your spend: it can route you to a frontier model and you pay frontier prices. Cost is the only mode explicitly designed to squeeze your token spend. If the reason you’re turning this on is the invoice, Cost is the setting, and it’s worth being precise about that behind closed doors before someone assumes Auto means cheap.
If cost per token is new territory for your team, What are tokens in AI and why do they charge you for them is the foundation that makes this discussion legible. And if you’re interested in the same problem from the gateway side, LiteLLM’s router charges two calls per request — with the difference that there the lever is yours.
But the real story isn’t the modes. It’s the admin panel. An administrator can enable the router per team or per group, restrict which optimization modes members can even select, set the default mode, and maintain allow/block lists for the underlying models. There are soft and hard enforcement options to standardize everyone into Auto. And by default, the routed model stays hidden: the developer doesn’t see which model actually responded, unless the admin decides to show it.
Hold onto that last one. The knob that lived next to the chat — what model am I talking to? — is now an organization-level policy decision, and the default is opacity. For a platform team standardizing spend across fifty engineers, that’s exactly the control you wanted: predictable costs, nobody burning budget on the fanciest model to rename a variable. For the individual developer who built intuition about which model handles which task, it’s a real loss of legibility. Both things are true, and which one you feel depends entirely on which side of the dashboard you sit.
Some data to plan with. It comes enabled by default on Teams plans: if you’re on a team plan, you might already be routing without touching anything, so check your mode before assuming your last week of usage means what you think. It runs on desktop, web, iOS, CLI and the SDK, so this isn’t just an editor change; your CLI and automation traffic routes too. And Grok 4.5 is required as one of the price-efficient routing options, which says something about the economics Cursor is optimizing against.
My read, from the CIO chair: this is something good, launched with a default I’d change on day one. The Cost mode plus mode restrictions is a genuinely useful governance primitive: the first time Cursor spend becomes something you can standardize instead of just hoping it works out. But hiding the routed model by default changes your engineers’ mental model for a neatness nobody asked for. If I deployed it, I’d leave Cost or Balance as default, and turn on “show routed model”. Let people see what answered them. Governance and legibility aren’t opposites, and the default shouldn’t force you to choose.
Already have it on without knowing? If you’re on a Teams plan, check what mode your Auto is in — and tell me if you’re going to leave the routed model hidden or visible for your team.