How to run Copilot Subagents in parallel?

[BOLT04]

Select Topic Area

Question

Copilot Feature Area

Copilot Agent Mode

Body

I was looking at these docs but could not find a way of running sub-agents in parallel. I only saw this "Subagents don't run asynchronously or in the background".
It would be really efficient in tasks like "deep research", or even sub agents tasked to search and explore the codebase to find a bug.

Is this possible?
If not, is there any way I could submit feedback for this to be a feature request?

Thanks in advance!

[JOYBOY-3]

No, sub-agents in Copilot Agent Mode cannot currently run in parallel — they execute sequentially. If you’d like this to become a feature, you can submit feedback through the Copilot app’s feedback option or via the Microsoft Feedback Portal.

[jakub-trzebiatowski]

A high-level agent (e.g. GPT 5.2 or another smart / more expensive model) could prepare the plan and give specific instructions to a low-level agent (e.g. GPT 5 mini or another fast / cheaper model), which is then executed in parallel (e.g. for all modified files in the Git repository).

A whole range of repetitive tasks could be solved this way. I'm not saying that it should be the default, but it would be great if Copilot had a user-selectable option to act this way or figure out that this is the right strategy at the early phase of planning.