I have a Pro subscription, but I’m having a problem with the usage limits for Claude 4.6.
I sent only about 3 prompts, and my limit was already exhausted. I was surprised because this happened extremely fast, especially since I’m paying for the Pro plan.
I’d like to understand how the usage limit is calculated. Does it depend on prompt size, context length, previous conversation history, or Claude 4.6 itself?
With only 3 prompts, I didn’t expect to hit the limit so quickly.
Could someone explain why this happens and what the actual limits are for Pro users?
Thanks.
Think of non-Gemini models as a “sample”. That quota doesn’t increase with your subscription and is expected to deplete super fast (enough to make three request with Opus). Please, read your subscription benefits.
From my part I wish they removed them, they are useless.
Cheers.
Thanks for the explanation. Do you have an official source confirming that the Pro subscription does not increase the quota for non-Gemini models? I checked the subscription benefits, but I couldn’t find any clear information about the actual Claude 4.6 limit or that it is only intended as a sample. My main concern is the lack of transparency about how this quota is calculated.
Being an Ultra Plan user for 3 months now, I can finally admit I am giving up on Antigravity, had the worst experience working with it. Thank you Google.
Renewed my ultra plan yesterday, made some few changes to my codebase, and was hit with a 3hrs wait for the Opus Model, only to return and after 3 exhaustive hours of wait, to be hit with the 6days wait, without making a single change to my codebase.
Probably, not recommending anyone to buy the ultra plan or even use Antigravity at all.
It was … while it lasted, switching to Claude Code