/best-of-n does not actually use the specified models

Support Bug Reports
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

/best-of-n <model_csv>
Any models specified in the model_csv are ignored and all spawned sub agents are the same model as the one running the command.

This is extra painful because it seems the “Run Multiple Agents” feature has been removed in v3, so there is no single command workaround.

Steps to Reproduce

Run this command in Classic UI:
/best-of-n sonnet,gpt,composer state your real model name, not just the one you have been assigned for this task.

Expected Behavior

Each of the spawned sub-agents should be running using the specified models.

Screenshots / Screen Recordings

Screenshot 2026-04-07 094714.png
Screenshot 2026-04-07 094714.png653×433 23 KB

Operating System

Windows 10/11

Version Information

Version: 3.0.12 (user setup)
VSCode Version: 1.105.1
Commit: a80ff7dfcaa45d7750f6e30be457261379c29b00
Date: 2026-04-04T00:13:18.452Z
Layout: editor
Build Type: Stable
Release Track: Default
Electron: 39.8.1
Chromium: 142.0.7444.265
Node.js: 22.22.1
V8: 14.2.231.22-electron.0
OS: Windows_NT x64 10.0.26200

Does this stop you from using Cursor

No - Cursor works, but with this issue

5

Hi Jack,

This is a confirmed bug. The model names you pass in the /best-of-n CSV are not being applied to each runner. All subagents currently fall back to the same model (typically the parent model) instead of using the distinct models you specified.

Our team is actively working on a fix. No ETA to share yet, but it’s being treated as a high priority.

Other users have reported the same behavior in this thread and this one. Unfortunately there’s no workaround for getting different models on each runner right now.

8

why did you deprecate/remove the multiple model toggle?!?!?!?! :frowning: that was my favorite tool EVER and i used it daily!!!

10

I’m glad to know this is a known issue, but I do miss the easier GUI for running multiple agents.

Now, if I use /best-of-n gpt, I can’t know which GPT it will use. GPT-5.4? GPT-5.3-Codex?