Cursor launches Cursor Router to cut model costs by 60%
Cursor says its new routing system picks the best model for each coding task, with team controls for cost, intelligence, and access.
Entities: Cursor Router
Cursor says its new Cursor Router feature can automatically choose the right AI model for a coding request and deliver "frontier-quality" results at 60% lower cost. In its launch post on X, the company described Router as an intelligent model selector for Teams and Enterprise customers, with an Auto mode that can optimize for Intelligence, Balance or Cost. Cursor's blog says the system was trained on more than 600,000 live requests and tested across millions more, and that early-access customers saw roughly 30% to 50% lower costs.
Combined views
278.8K
12 posts, first seen 1h ago
Reactions from ranked influencers
12 posts
Select Auto mode and whether you’d like to optimize for Intelligence, Balance, or Cost. Cursor Router analyzes each request and routes to the best model for the job: frontier models when the work demands them and price-efficient models when it doesn't.
In early access, customers observed no drop-off in quality with a lower cost per commit vs. routing all requests to Opus 4.8.
Introducing Cursor Router, our intelligent model router that selects the right model for the task at hand. Router delivers frontier-quality results at 60% lower cost.
Model routing is a first-class feature now.
Introducing Cursor Router, our intelligent model router that selects the right model for the task at hand. Router delivers frontier-quality results at 60% lower cost.
Makes so much sense for Cursor to build this. Getting the “intelligent” part of intelligent model routing is going to be key, and I’m excited to try this out
Introducing Cursor Router, our intelligent model router that selects the right model for the task at hand. Router delivers frontier-quality results at 60% lower cost.
@cursor_ai nice!
We desperately need a smart model router. 1. We’re seeing a model explosion: GPT-5.6, Grok 4.5, Muse Spark 1.1, GLM-5.2, and Fable 5 all launched within the past month. 2. Even for a single model family like GPT-5.6, there're 3 (Sol, Terra, Luna) and 5 reasoning-effort levels. That is far too many decisions for users to make manually. The best model should be selected automatically based on the task, latency, quality, and cost.