Hi, I’m an Antigravity Ultra user. After the latest updates, my quotas started running out much faster even though I’m using the same prompts as before. I usually use Claude Opus, and it would typically only drop to 80% after about 2.5 to 3 hours of use. Now, it drops to 40% in less than an hour.
Is there an ongoing issue? Honestly, I couldn’t figure out what’s causing this change.
Thanks in advance for your help! <3
According to the mods here:
To provide you with greater control and a seamless path to scale, we are evolving our Google AI plans by introducing built-in AI credits that can now be utilized directly within Antigravity.
Under this updated structure, our Google AI Pro tier is tailored for practical builders, offering generous limits for Gemini Flash alongside a baseline quota to experience our most advanced premium models. For developers operating at the highest scale who require consistent, high-volume access to our most complex models, Google AI Ultra serves as the ideal comprehensive solution.
Furthermore, we’ve designed these plans with maximum flexibility in mind; if you are on the Pro plan and require additional capacity or deeper premium access for intensive development sprints, you can seamlessly top up your AI credits to customize your limits and continue shipping your best work without interruption.
But seeing that you’re already on the Ultra plan there is more that they are not telling us
Exactly. Even with a very basic worker project, my quota immediately dropped to 60%. All it had to do was analyze a simple worker setup and write a new one to make an API request.
In the past, I used to run much more complex tasks—analyzing backend, UI, and multiple other projects simultaneously—and the quota wouldn’t even budge from 100%. To be honest, it’s starting to feel like we’re being shortchanged here. Since Ultra is already the highest tier available, I don’t see what else they could possibly try to upsell us.
It feels very inconsistent and, frankly, quite suspicious.
“you can seamlessly top up your AI credits to customize your limits and continue shipping your best work without interruption”
But many users have said that doesn’t even work and the ones for who it works burn through credits at accelerated rates
Mmm. Before paying per use, it would be nice to have an analysis and monitoring of token usage within Antigravity, so as to understand how much tasks cost and make predictions, rather than waiting for the charge to be updated not even in real time but 24 hours later… it’s not profitable/predictable from a professional perspective.
Am I wrong?
You are absolutely right. So, what are our options here? Am I supposed to just wait for 5 hours? And how can Google just change the rules on a whim like this? I specifically subscribed to the Ultra plan because it was supposed to offer the highest limits available.
Does this mean a Pro user’s quota only lasts for a minute now? It’s ridiculous.
I am also on the Ultra plan and I have experienced a similar issue. Since this week I have been exhausting my quota on Claude after only 1 or 2 hours. I used to work with 4 or more agents running at the same time but now I can hardly run 2 instances before my quota is completely depleted.
This behavior is very inconsistent and switching to Gemini models is not a viable option since they constantly hit the famous Agent errors reported by others in these forums.
I actually went from Pro to Ultra to get the full capabilities but the development experience with Google Antigravity is crashing and no longer provides a consistent flow.
Same here, and the numbers are hard to ignore. Before the update, running multiple agents simultaneously on complex tasks for 5 hours wouldn’t even bring me below 80%. Now I hit 0% in under 90 minutes doing the same workflows.
That’s not a minor adjustment — that’s a dramatic reduction in what we’re actually getting for our money. And like others have said, there’s no transparency about what changed or why. The “top up your credits” response isn’t an answer, it’s an upsell.
If the Ultra plan no longer delivers what it used to, that needs to be stated clearly — not hidden behind vague policy updates.