Are they doing something internally or swapping models or something? I used Gemini 2.5 Pro 2 days ago and now when I’m using the same model it stays “reached your rate limit”.
From what I understood, 2.5 Thinking has limit of 2 RPM/50 RPD but not able to understand how I have hit the limits when I even not used it for a whole day and starting fresh now. Is there some limit in per prompt window also? because it works fine in new prompt.
no guys, it doesn’t. just confirmed now. 
Same issue – even new prompts now show ‘rate limit reached’ despite minimal usage. Confirmed it’s not working in fresh sessions either. The documented 2 RPM/50 RPD seems miscalculated ?. Could there be unannounced model-side throttling? Would appreciate official clarity on how limits are enforced…Thanks!
And when I ask why it is in the other flash versions it says "Just a sed…) and keep whirling!
Also experiencing this, no idea why. Between this and the CPU usage / extreme lag, it feels like something was broken in the last few days.
It does seem like happening for past few days only. Never faced it for basically any model before. Don’t know how come 40k tokens are freezing the whole UI and lagging like crazy now.
@Corkster52 did the same, got nothing reasonable from it.
Did anyone get a reply out of this? I’ve been having the same issue for the past 3 days now
I’m having the same problems and also for about the same time. I guess Google may have underestimated demand on Gemini 2.5 Pro. The claimed free rate now is 25 req/day but I’m getting barely 5
Same issue, hoping it gets resolved soon
Hi @mgks ,
Welcome to the forum.
Thank you for your feedback. We will escalate this issue to our internal team and will get back to you.
Same issue today with Gemini PRO plan (Europe, Czech Republic)…Last two days a trired create 2 easy Canvas wep apps. …Now for image generator I using my Canvas app, with imagen-3.0-generate-002 thrue my API on Google Cloud. …Is possible that this API image generation could take “more” resources from my Gemini PRO plan than If I generate this images thrue chat? …I dont understand the limits for PRO plan. We need some overview of drawing resources of our PRO plan!
this just popped up for me as well. when i asked gemini to create certain reminders for me vs just speaking with it.
however, the learning coach chats i have with it still work just fine. it just seems to be the flash versions that it is saying now. its saying i have to wait til 11:10pm tonight to refresh. however, given that i have been using it for weeks now, and this is the first i’ve had it happen, i’m starting to wonder if its because i exceeded its limit by asking it for reminders and things to persist through to other days, not just in the moment responses.
I’m getting this now - i don’t think i used all my submits for the day, very frustrating. I upgraded to the paid tier about a week ago, never had this issue while on the free tier. Don’t know what do - may have to try the anthropic platform i can’t wait for this to start working again.
I just want to say, that I have the same issue since some hours.
I’m in the free tier and had never any issues with “rate limits” before. I also switched to the flash models, but it’s the same issue.
EDIT: I just tried the Gemini 2.5 Pro model again and it works now. 
Mine’s still not working - so frustrating. I just tried to use Claude and that seems unworkable, i may try Open AI I can’t just stop developing for the day that’s not tenable. This never happened before with many months of using AI studio.
I have the same issue. Now I need to stop developing for today… unbelievable, NOT VERY PROFESSIONAL. I saw there was an announcement for Premium with higher limits, but I can’t apply… it say already PRO???
Same here It just says “You’ve reached your rate limit. Please try again later.” and doesn’t work. I’ve been using it for a while and there is no way I actually reached the limit this early today.
It has happened to me today. Really weird
Nvm, it was back after 1-2 hrs. Seems like such limits tend to be temporary and perhaps unintended, IDK?
i just got this same message today 07/19.