How to Choose and Switch AI Models in Cursor

Cursor lets you pick which AI model handles a request, and the choice matters: some models are faster and cheaper, others reason harder on tricky problems. This guide shows where to switch models, how Auto mode decides for you, and a simple way to match the model to the task.

What you need

• Cursor installed and signed in
• An active plan or trial that includes model access
• A task in mind, from a quick edit to a hard refactor

Step 1: Open the model picker

Open the chat pane and look at the bottom of the input box. The model name sits next to the send button. Click it to open the picker, which lists the models available on your plan plus an Auto option.

Step 2: Understand Auto mode

Auto lets Cursor pick a model for each request based on the task and current capacity. It is a good default for everyday work because it balances speed and quality without you thinking about it. When you need predictable behavior, pin a specific model instead.

Step 3: Match the model to the task

Use a fast model for small, well-defined edits where you already know the answer. Reach for a stronger reasoning model when the change spans many files, involves tricky logic, or you are stuck and need it to think.

Step 4: Switch mid-conversation

You can change the model between messages in the same chat. If a fast model gets stuck, switch to a stronger one and ask it to continue; it keeps the conversation context, so you do not start over.

Result: you spend reasoning capacity where it pays off and keep simple edits cheap and instant, instead of running everything through the most expensive model.