Copilot Pro plan, Agent mode limit

[metalurgus]

Select Topic Area

Question

Copilot Feature Area

Copilot Agent Mode

Body

I have a Copilot Pro billing plan.
On the plan description page it says:
Unlimited agent mode and chats with GPT-4.1
With a footnote:
The base model at the time of writing is powered by GPT-4.1. This is subject to change. Response times for the base model may vary during periods of high usage. Requests to the base model may be subject to rate limiting.

I am using Agent Mode with GPT-4.1, which supposed to have no limits in agent mode.
While I am sometimes encountering rate limiting, which are described in the footnote, and as I understand - happen during escessive load, I am also getting errors like this:
Sorry, you have exhausted the agent mode usage limit. Please switch to ask mode and try again later.

My plan literally says there is no limit, and yet they are limitting me.
What is it? False-advertisement? Or maybe I don't understand something?

[rodecuir]

Yes it states "unlimited". However, there seem to be secret limits on all subscription plans, i'll refer you to this discussion: https://github.com/orgs/community/discussions/148896#discussioncomment-11803831

[metalurgus]

Thanks, I have read it now.
There is another message I am getting for rate limit.
It says something like "Sorry, but you have been rate-limited due to excessive load on %enter model name here%...", and I believe rate limitting is related to specific model usage, while my error message "Sorry, you have exhausted the agent mode usage limit. Please switch to ask mode and try again later." appears for every single model I try, and actually says it is about agent mode, not specific model

[rodecuir]

There's supposed to be a usage report of your requests at https://github.com/settings/billing/usage, it is supposed that with Pro+ subscription, you have an allocation of 1,500 premium requests per month.

Premium request usage is based on the model’s multiplier and the feature you’re using. For Copilot coding agent uses a fixed multiplier of 1 for the premium requests it uses, and may use multiple premium requests in response to one user prompt. ↩

I just found also this discussion on rate limits, maybe it can help: https://github.com/orgs/community/discussions/160829

[metalurgus]

Nothing shows in there for me (usage report), for neither period...
Also I am using just Pro, not Pro+ (at least not yet)

[rodecuir]

Then i hope the discussion gets to someone from GitHub, I couldn't find anything else...

[metalurgus]

What a coincidence, but next day after I opened a support ticket with quite the same text as this discussion, agent mode limitations disapeared, and I did not reach agent mode limit a single time for the last 3 days ;-)