#general・🧺
1 messages · Page 36 of 1
So does opal
Me too
You do know these are experimental right
Programming is not useless, it is automated now
So you don't have to focus on code, so that u can focus on problems
See there's no google, and openai has a strong tie with nvidia
..
Google is not a dead company, it's a trillion dollar company that's trying to catch up, and if u talk about stocks, idk nothing about it
I do know a little, the basics about stocks
I know how they work but I dunno what to buy
They are in the smartphone market, the ai market, and the search engine, bro forget it
You ever use bing instead of google?
use brave
That's where these guys come
Brave is a browser I think
ya
Ai doesn't suck, u haven't given custom instructions that suit you
- os
Ya
Android
I bet if google doesnt out of business within ten years, you have to buy every single thing from Google, challenge?
chrome os , chrome os flex , Aluminum OS (Aluminum OS is not launced yet )
-_-
They may be fragile, but imagine if google disappears overnight android shuts down, 72 percent of phones become e waste, millions lose jobs
It cant happen
They won't let it
They may be underperforming in stocks
Hmmmmn
android wont
its an os on phone
You use android 5 now??
but there can be any other os to replace it
qualty of updates
Ya ik the updates are trash sometimes but c'mon, android 17s gonna be great
hmm
we all are
What about microsoft then
Slapping ai on everything just to be called microslop
I've seen users get banned from copilot server if they type microslop
Ai integration is not the greatest idea
really
agreed
Actually there's even a GitHub repository to remove all the deep system ai integrations from windows 11
hmm
they will fix it
We don't need new system, we need a killswitch to toggle between ai integration and zero ai
Which they obviously wouldnt do
not posible
That's what I said
That means you understand the minimum requirements of the system will skyrocket
Cpus will need an npu
And better integrated graphics
Instead I could call you directly
Oh
That's true
Ofc
Ukw, human intelligence is powerful than ai
GPUs in large ai data centres consume at tens of thousands of watts which is enough to power a small city
And it operates 3000-3500 TOPS
No
Their learning is slow
And fixed
They can never beat human intelligence in this timeline
Our brain just needs 20w power
Our brain operates at hundreds quadrillions of synaptic operations per second
Ai doesn't forget things, we forget things, that's the difference
The guy who founded click to prove u r not a robot??
Oh
Sounded familiar
Anyway let's move #off-topic・🐢
Or the mods will get mad seeing our texts
Relatable
@outer wagon add a spam protection bot like wick or dyno
Hi everyone.👋 Curious on your thoughts on the evolution of normal Gemini Gems? Do you believe they will be autonomous one day or something else like Opal would replace them? It would be amazing to be able to convert my original Gemini Gems into something more magical.🙂 – Although it seems like I need to recreate them in Opal and hope they become more autonomous. Would love to hear other people's thoughts on this.😊
sup
thanks gemini i made my own rpg game. it require massive code but easily made it 👍
Hello! I have a question about Whisk. Yesterday, I had access to many different image formats while creating photos. However, today I can only see three formats. Does anyone know how I can recover the other options? Thanks for your help!
.
did they nerf gemini in antigravity recently ?
Hello! I’m pretty new to Flow, and I’ve started noticing that after a few x2 upscales, I receive an error message and have to wait around 24 hours before I can start upscaling again. Is there some sort of limit for this?
anyone wanna play the google popcorn game? https://share.google/7iEl86dDMD8wxb3b6
Hello am new here
😑
WHAT DID YOU DO TO NANOBANANA PRO?????
WHY IS IT NOW REFUSING TO MAKE AN IMAGE OF A SIMPLE, GENTLE KISS?
WHAT HAPPENED?
I was trying to simply make a pretty tame kiss between two adult, consensual characters, in a cartoon style!
How much time did it take
I guess they made it sensitive instead of sensible
hii
all functional UI/npc/enemy/field map/town/boss/mob/skill/pet mount/visual sprite it took more than 3days. i add little by little 3 to 5 hours a day
What do you think is going wrong when Permission is denied when editing my image with NB on https://aistudio.google.com/?
My image is not explicit and retry won't work
Hello 👋🏻
tbh I think the next big narrative for stablecoins might actually be credit markets. Capital is slowly moving there already. PACT’s idea is basically letting stablecoins run through the whole lending lifecycle, instead of just tokenizing assets. Kinda interesting approach. btw saw Gate running a CandyDrop for it, trading more gets more rewards and max is around 640K PACT, ends Mar 9.
Cool
I've seen rpg games take a month even when working 24/7 back before ai came
hi
Just posted my AI Action Short Film : Anyone interested in watching movies can watch it and let me know your feedback on how it was
This entire action scene was created using Nano Banana Pro and Kling 3.0.
No camera. No crew. No set.
Just a vision — and the right tools to bring it to life.
Watch till the end. 🎬
Generated using Nano Banana Pro & Kling 3.0
Prompts, direction and creative vision by aieyestudios
💬 What do you think — does this feel real to you?
👍...
are you just a troll. a university student that can't even do his own research 😐
My source is i made it up 😛
post on flow instead yo. might get taken down for "self-promotion" rule 🙂
i hope opal replaces it. but opal needs to improve a bit 🙂
Thank you.🙂 How are you currently using it? Have any good inspiration examples for me to try.
did they dumb down gemimini in antigravity in the past few days?
Does anyone know if there is a Gemini gem Discord?🤔 I have a large free gallery I made to help beginners get started.
Yes
there's a gemini discord
Could you share the direct link? Thank you.🙏
DM
dark holer
hello guys
since the new update i'm unable to edit the ingredient archive. I want to delete some images, but they are... there...
help?
its there any free server provider?
Promoting is not allowed here
who are you?
According to him, he's you but stronger.
Just a troll. Ignore and block.
thnx
According to him, he's you but stronger.
Why did bro start a thread for that 
I made fun rpg game update wizard class 😁
Hi
hi :)
i just been re ordering my spotify
yeah they have a special 3 months free even if you're a past customer
damn maybe i should unsubscribe and see if they'll offer me that 😂
they might
rn i'm experimenting with a LoRA i trained for flux2 klein 9b, i tried some gta 6 screenshots but that game is so realistic it hardly made it better. really can't wait for it to come out
Oh yeah? Nice
ty :)
Proud of you
i appreciate it!!
it was trained on 5000 mathematically sorted 8k images with permission from unsplash and 16000 steps
supposed to make everything generally more detailed and fix small annoying inconsistencies in ai generated images, but since flux2 klein 9b is also an image editing model that takes a prompt, it ends up working even better
what happened
Lol nothing just my vape popped juice into my mouth
oh no 😭
That's awesome though man
ok right now i have a small problem, it really likes making everything yellow 😂
before/after
Super Nice
Well the clarity may seem lesser but to be honest I think the yellower version captures more "soul"
Sweet.

ah i know why that is, the output image is always exactly 1mp so in this case it's 1024x1024px and the input may be higher (here it's 2048x2048px)
but i can change it to match the input resolution as well :)
Makes total sense
i have tsv's of 5 million images but i lack a 40tb hard drive to have the images on 😂
rn i'm limited to 1tb lol
Can you easily go back to the "clear" vrrsion
yeah i save both versions so i can just pick whichever is better
I only have one beefy boy HDD with 12TB. That's a lot.
True. That's really helpful
especially in these times 😮💨 i looked at prices to see if i could store it locally but yikes
the 5 million image/text pairs takes up like close to 100gb and that's just pure text
that's so cool :)
me too. can't wait for it to finally get hotter and brighter in the evenings so i can stay out for sunset
You can, but you'd have to use HDDs.
definitely. even then i think it's gonna be expensive haha
So I suppose it cannot be a floppy disk? 😛
It could be
worth trying hahaha
If you're patient, you can get refurb HDDs for less than $10/TB. Before I got my big boy for cold storage, I had a 6TB refurb for $54 bucks and put it in an enclosure.
Aye noice
hey that's actually not bad! not compared to an m.2 nvme ssd
Hi all, I have a question regarding google Gemini and NotebookLM. It looks like I can't add notebookLM to my Gemini Gems anymore. I can still add it to a conversation but not to a Gem. Is this something that has changed? A few days ago this was still possible...
People who buy big HDDs usually want the big ones, so the mid-range 4-6 TB refurb drives can be cheaper per TB. Find some popular refurb listings and set one of those price tracker things to email you when prices drop under a certain level.
Oh
I don't, but NotebookLM has their own server. I'd check there if nobody here knows: (won't let me share the link)
i could totally split up the 40TB between HDDs :) they're all images and they're like 2-30MB each. thanks for all the tips
Wow
i don't plan on using the text pairs that follow, they're all user defined captions and some are literally empty lol
yo wait did it get better or worse? new one is the right one
oh i'm purple now :) epic
I really still like the left one
same ngl
yeah but the details kinda get lost
the original purpose of this was to turn any render into a photorealistic image
I figured
Yeah definitely see what you're saying
I wish I understood half of what you do lol.
Or maybe even 1/13
don't worry i do stuff and then i try to understand it xD
a really funny side effect of the LoRA is that it'll take any image (even memes) and make them hyperrealistic and creepy
right xD
Wait
afraid to test it out on cartoons, idk what will happen lol
whats up
Oh sorry was trying to DM someone and sent it here lol
ah fair :D
it happens
final image because i feel like i'm spamming at this point lol
there's lore behind this one: the left version was originally an image of my desk that i edited with nano banana hours after it came out for the first time. started by replacing my ceiling with grass and then asked it to add as many cats as possible.
on the right my LoRA along with flux2 klein 9b edited it
How's that going
Has anybody come up with a creative AI use case that integrates looker for either its semantic modeling top optimize comprehension and accuracy, or a an interesting end to end use case with the mcp for Looker?
I'd love to see Gemini interpret the works of Scott Joplin. I know music rendering is still fresh but hoping gem gets it nailed
I find him to be quite a superior composer, it makes me feel almost sick lol
I'm new here
No comment
I mean if you feel that way, it is probably for the best. Mental health is important. Also, I hardly use Discord
What does it mean if I spend my day stressed out then?
lol
Wdym I wasn't talking about AI
If you have constructive feedback, I am sure they will listen to any suggested improvements you might have
Trolling isn't funny and it is a waste of time
what is it about Discord where the trolls always come out at night?
Some guy sent the n word a couple min ago
Hello! Just a friendly reminder to keep our discussions focused and to avoid excessive emojis, spam, or content that doesn't make sense. We've deleted the recent messages that didn't comply. Thanks for helping us keep the chat clean and engaging for all members!
Thank you CatShire!
Burn it all down and go back to telnet prompts and Usenet! haha
BBS? Those were pretty neat
hi there i havew a question about antigravity paid plan usage, after one day of 20usd payment, it keeps locked the claude and gemini pro models for one week, is there smth i am not getting, I'd be glad whoever helps me with this issue
good day guys
Have a great weekend ahead
Hey, how are you all? Something strange, for the first time, happened between me and Gemini today. It went totally rogue on me. Although this is some kind of a prompt that I do on a daily basis, I do it five times a week, but today the answer was extremely rude. It felt like there was a glitch or something. Of course I took screenshots and I kept the conversation but did this ever happen to anybody else?
nah thats new its usually helpful but goes off topic alot these days i rlly don't know why
like even when im using the gemini CLI or antigravity it solves problems in ways that are not neccessary and just wastes tokens
Hi everyone!
there is any way to get pomelli API key to implement it over my platform to my users?
There is something I don't understand. Over the years, technology and machines have taken jobs. For example, in the Industrial Revolution, machines took jobs. But isn't that what AI is doing today? Why do people hate it more? Is it because of the environmental concerns, even though I feel like the Industrial Revolution polluted the earth more
Sometimes it's just about empathy, so I guess just wait till it happens to you and we'll see. Tho yeah, in this new world gotta learn to adapt fast, those same tools can now even be used against themselves if you took the time to learn what you're doing.
Pretty sure I'm actually the one who took some bot's jobs here and there lol.
I'm not wishing for it to happen to anyone, because I wouldn't want it to happen to myself. I don't think people should be fired; AI should be used as a tool to help the people who have those jobs. not take the jobs. To be honest, most of the time when companies try to replace people with AI, it backfires, so it is just better to let them use AI as a tool, instead of replacing their jobs. But since AI helps people work faster, the company might still have to cut off some people.
One thing I feel like AI might not really replace is human creativity. I feel like humans are way more creative than AI.
Ai
hello i'm a new random french guy and i'm happy to be here
"AI should be used as a tool to help the people who have those jobs. not take the jobs."
Giving people tools that make it so one person can do a lot more work is the same thing as taking jobs. You can't have one without the other.
True, thank you for pointing that out. But what can we do:/
why are you happy to be here
using AI is also a skill isnt it?
I believe you can achieve better results by using it as a tool. For instance, understanding how to prompt effectively is key to getting the best answers.
reference of the film I am the legend
exactly the better you prompt the better it works. whoever is employed should learn things around their domains like how they can leverage AI in their work.
reminder, general is not the place to ask questions about products as we have #flow-general・🌊 , #whisk-general・💭 , #jules-general・🦑 , #opal-general・✨ and many more channels where your product specific questions can be asked. Thank you
I've seen someone prompt AI so good that it was basically an essay. The difference between the basic prompt and the advanced prompt was big; the advanced prompt gave way better results.
<@&1366831844502274180>
Hello! Just a friendly reminder to keep all conversations in English for this server. This helps our entire community engage effectively, and we've gone ahead and deleted the recent posts that didn't comply. Thanks for your cooperation!
it will be an essay if its a good one lol
did they remove the delete feature on flow extend? i cant get rid of any clips in extend
because i want have new experience and how use opal
Top of the Morning to ya all
Hi!
Hello fammyyy
Salut il y a des Français ici ?
Anglais. It's an English only server.
Where can i report bug?
good day mate
Tag mod and share to him
Have a nice day sir
I feel really bad ,that I need to wait for what I wanted to type
Hello
Hello
Great Morning to you all
Hello everyone
Hello
hi
Wassup
good morning whiskers
...
hiii
Hello all 🙂
Hi to all the people that just come in
<@&1366831844502274180> The censorship with nanobanana on Flow is getting out of control. Even for a simple prompt like "Hi!" you get this. When I do x4 generation, only 1 image is generated, the rest 3 is marked as going against guidelines
hi! unfortunately mods can only assist with moderation/ discord related issues. we recommend using #1238621511401734254 to submit any bugs and feedback to the labs team. these threads are actively monitored by the team and so may be seen quicker than in a general chat!
um new here , is the asset option ain't supposed to work ?
Hi there! Please use #1238621511401734254 to submit bugs and feedback to the labs team. Unfortunately, we're only able to assist with moderation & discord issues. Thank you!
Hello, i have a question, What's going on with anti gravity? there is no way I used my quota? it just hit limit instantly? i asked 3 questions with images so it can verify the work i did myself? It hit on claude and other models as well.
hello
use whatever tooling you can to find out how your quota was spent on the individual parts that make up your actions/tasks. I assume you accidentally underestimated some thing's expense. Or possibly your account type has very low limits, but if you're paying I don't expect that to be true.
hey! welcome!
😑
How do i find the account type limits? it doesn't say on settings here.
Hi there, we do not have any antigravity channels here. The best place for antigravity currently would be the Google Developer Community. There is the vibe-coders channel. Here is the server invite. https://discord.gg/google-dev-community
Man, Discord giving a lot of trouble to load in many locations atm. Not even gonna ask if it happens to someone else cause more likely those having the problem can't even get in.
Discord just had an outage! looks like they're fixing it
https://discordstatus.com/
Welcome to Discord's home for real-time and historical data on system performance.
^that was my first successful message today! haha
Yeah I'm actually here cause it only worked from Japan
Why is pomelli not available in eu?
And the problem is only affecting some servers, not all Discord in general, some are working like nothing. It'll fix itself.
It isn’t just you. I have been having trouble all morning and afternoon. 😭
We'll just wait they'll fix it.
And like you just said just some servers
I didn’t see that till after that posted, that’s how bad it is.
Was going through the same, Discord's going like if nothing's happening from a Japan VPN tho.
Kinda feels lonely here atm tho.
The recent announcement does not convey at all what Pomelli is or why someone should want to click on it, btw.
Probably a good idea to always be reminding people what these codenamed projects do for them, IMO.
thanks for your feedback! [we announced Pomelli](#labs-announcements・📣 message) back in October, with this announcement being an update for folks who opted-in to the Pomelli role for updates, but will keep this in mind for future announcements!
Ahhh, it's for marketing. 😅 Explains why it wasn't on my radar.
hi i am new here
welcome!!!!
hi
no worries 🙂 good shout to remind folks, especially new members. thanks for this!
I got to get some teeth worked on tomorrow🪥
Can Nano Banana 2 understand proper sizes, measurement, distance, and other real-world physics application?
I'm trying to use real-life measurement to describe an object
hey @outer wagon sorry i have a question but i don't think i can ask it in a public channel what can i do?
feel free to DM <@&1447643696152576081> to talk to a mod!
Yo
hey guys! anyone know if there is an MCP for nano banana? I would like to allow my claude to generate assets using nano banana..
hi
why is pomelli not available in UK?
why gemini canvas got error all the time, when gpt 5.1 and sonnet 4.6 can fix error flawlessly..
I have been using Veo3 for months and am usually amazed. But today I have been getting garbage results. Dialog not staying connected to a character, people walking out of walls and duplicating.
Here is a prompt that I used along with two ingredient images. There are two big problems. It started with one ingrediant as if it was a start frame. Then it created duplicate humans.
The robot robobob
walks up to the fredhouse
and says in a friendly computer voice "Is Fred Turner here?"
A man wearing dirty blue jeans and a sweatshirt walks out from behind the house. The man asks "What do you want?". His voice is suspicious.
The robot says "I am your new robot."
The man looks surprised.
There is no other dialog.
Here is another one. I was more specific in my diections.
The scene open on the fredhouse
The robot robobob
walks into the frame and says in a friendly computer voice "Is Fred Turner here?"
One man, Fred, wearing dirty blue jeans and a sweatshirt walks out from around the back of the house. The man asks "What do you want?". His voice is suspicious.
The robot says "I am your new robot."
Fred looks surprised.
There is no other dialog.
I was able to get rid of the white opening frames but the dialog is not staying attached to the characters.
good day bro
Has anyone here tried multiple Labs experiments? Which one feels the most useful so far?
Any date eu/sweden?
Hello 👋🏻
Heyo, did anybody here ever compare aider to opencode to antigravity?
hi
Hello mate
??
hello
we have the labs community team and Moderator! there is also the Google team on the server!
is there anything we can help with?
I need access why have I not been given access?
Is there a way to use nano banana pro instead of nano banan 2? The new version is huge step back compared to old one and cannot be used for professional purpose
Yoo
yeah
welcome to real world
Ello
Hi guys, how i can use some of the google labs products, without beeing in the US? Some advice?
where a get drop q about antigravity guys ?
nah I am good man
Nano Banana pro is always at the drop down menu from Flow's image model selection.
But if you meant the Nano Banana from Whisk that's just called "Nano Banana" (Not pro or 2) wherever they have it, or "Gemini Flash 2.5 image" or simply Nano Banana 1 and Flow did have it before, not there anymore.
hey there 🙂 there are tools across Labs that are available outside of the U.S.! what tools are you interested in trying?
@outer wagon I dmd you my email like you said, can you verify it for me plz
Hi @dark rose Welcome to Google Labs official discord server
We’ve forwarded your message to the appropriate channel. Please check out our #server-directory・☎️ for our list of channels and their uses. We cant wait to see your shared creations!!
Pomelli and stax
awesome! Pomelli is available in 170+ countries, you can see the full list here. Stax is still in experimental phase, but with hopes to expand in future!
We’re working to bring Pomelli to as many people as possible. See below for a list of supported languages, as well as countries and territories where Pomelli is currently available to users ages 18+.
Pomelli is not yet available
Pomelli by Google Labs is currently not available in your region.
Where is Veo 4?
sorry i should be able to see the disco channel but i am unable to, what can i do?
veo 4 should be better than sora 2 and seadance 2
Can someone @ me?
All from one photo, not bad.
I don't know why but the Veo gave us receeding hairlines 😂
Being underwater is next closest thing to weightlessness.
I like the music that got added
@primal thistle
why veo 3.1 can't make fight scene? it still awkward.
Yes 🙌🎇
Hahaha, I should have thought of that instead of prompting something as boring as winking xd
Veo also found out I am asian
But I didn't have a diving watch and I also don't have arms that hairy haha
how do i talk or contact support for the "Google Studio AI" website to give them a bug report?
@clear sinew If you go to the client and click settings (its at the bottom above your email address) you'll see an option to Send Feedback. You can also try the Google Devs server here https://discord.gg/google-dev-community
Thank you so much!
OMG labs.google is so fascinating. Why didn't I know about this before? 😭
good day guys
GUYSS I GOT THE MOST EPIC IDEA EVERRR
Anyone having trouble upscaling video to 4k (with an ULTRA account)?
(4k images is workiing Ok. Only 4k videos is giving me trouble...)
yea i had that trouble
so i removed m ultra
i mean i didn't renew it
after it expired
that bug will not get fixed untill the google devs actully work on it so its better not to get ultra
It's been working Ok since today.
I also had the same problem a few month ago. Then the issue stoped the next day.
I sometime managed to bypass the issue by regenerating the video (ex: using the remove edit, to force a new version without actually removing anything), but this trick doesn't even help this time.
hello guys just joined, looking forward to learn from yall
I like Google Flow. But I Can USE ONLY ENGLISH IN MY PROMPT!!!!!
I made this using web assambly emulator JS on the google gemini canva, and the posibilities are infinite, you could make some frameworks to do the AI play the game. Today I see a tweet of someone telling that Chat GPT 5.4 can run Mario of NES and Gemini 3.1 can run TEKKEN 3!!! OF PLAYSTATION 1!!!
(For my defense... this was an stricly case of study and experiment, if you are a Tekken PS1 Dev, Im sorry...)
reminder, general is not the place to ask questions about products as we have #flow-general・🌊 , #whisk-general・💭 , #jules-general・🦑 , #opal-general・✨ and many more channels where your product specific questions can be asked. Thank you
there is no place to Gemini Canva product 😶
<@&1366831844502274180> pls ensure that no one sends too many links
I think the moderators members are already working hard on this
Hello, does anyone have any suggestions on a AI-based UGC app that I can use to promote my app? Preferably free and easy to use
Can anyone tell me what's the difference between gemini-3.1-flash (NB Pro) and gemini-3-pro (NB 2). I was wondering which one is actually the more advanced? Especially when I turn Thinking to "high" when using NB 2. I don't mind to loose some speed, I just want to make sure I get the best results possible. Would be awesome to get feedback from the google team.
why is google labs locked in europe ?
<@&1173646484311384114> hi guys, does everyone who subscribe on Google Pro having limited prompted issue ? which usually able to prompt until 100 now only between 20 - 30 prompts ?
do we have any update on "Upscaling Failed!" error on flow? it's been more than a day now
Which Project was it that is used for developing branding for a company?
I forget which one it was. @ me if you respond. Thanks
Whisk API coming ever?
hey there! [we're investigating this,](#1481161292118233158 message) thanks for your patience
[Pomelli!](#labs-announcements・📣 message) now available in 170+ countries.
hey! I'd try the Geminin community for this discord.gg/gemini
Any labs news has been no new project in while that is not us only?
Thank you this was exactly what I was looking for.
@acoustic cipher
I want to make the same thing exactly but with my logo
Good night
yeah
can u play a google game with me ?
what google games and sure
pop corn
i am on it
ok press start
i did
i joined ur link
hi, please be careful of links being shared on public platforms such as Discord!
wdym
ok bye
so i am suspect of bieng hacker
is there any lab good for studying? im from poland
can you create sound effects with google ai studio?
what happened with Antigravity? I had the quota bar almost full and then i see this on the model tab 🙃
Hello everyone
Good evening all
Well, it's morning here.
kool
Hello, is there any antigravity support discord,
cause when i use ssh, many of my conversation is missing from the previous day
<@&1366831844502274180>
same here
lf moonseye tome
have a nice day mate
oh yeah dude i did one feature change and it took away 40% of my ai credits
if this is how antigravity is going to look like for the future i can promise you google can add it in advance to their already long list of failed products
Hi everyone! I'm Daniel, a full-stack dev based in Singapore. Great to be here!
I’m currently architecting a real-time notification system for a web app and looking for something production-ready and stable. I’ve looked into Firebase Cloud Messaging (FCM), but I’m curious if there are better alternatives for purely in-app real-time updates (versus OS-level push notifications).
The Use Case:
Real-time alerts while the user is active in the web app.
Needs to be highly reliable and scale well.
Low latency is a priority.
If user did not check the notification after valid time, send email to that user (Optional)
Hey guys,
do you also have problems with Antigravity? I have the Developer Plan and I didn’t even use Gemini 3.1 Pro, but I immediately got a 1-week lock because I supposedly have no tokens left, instead of the usual 5-hour wait.
Why could this happen?
Hi guys
How to follow the announcements tab
Like will the admin give us permission or anything like that because on one person is continuously messageing in it
That Georgia
Pls tell me what to do
Thanks
Is it just me, or have antigravity quotas got worse? Since doing a bit of vibecoding for my project, I've almost reached my weekly limit for 3.1 Pro and claude.
Using the pro plan
Hello everyone 👋
can i ask something about antigravity? how i can make the agent panel look like cursor and make it flow while writing down everything
same lol
the token limit are live updating (its not stable OR FIXED) so you could directly lock for one week or 6 hour wait time based on your luck
the limit are low as some point.....
I think its something with a weekly limit and 5 hour rolling limit so it could be that you alwys just go to like 20% before it reseting but it still countting to weekly
I'm having issues with uploading images to flow
but i hohenstly dont know
For claude models
I got ONE 5 hour limit and the next are always 1 week
I'm student pro plan user so ig they just limit more
For gemini 3.1 pro cause i use low version once i have always around 2-3 time 5 hour limit reset and the next are one week again (yes)
I dont really use ai credit so i just use it on antigravity lol
There is an AI credit?
i thought it was all seperate
also I know why my limit is gone.
hello, i recently started using antigravity and wanted to know how to actually use claude opus latest model. i ran out of free usage and am confused as to how to actually use claude, if i had a subscription for exemple.
should i use the plugin in antigravity ? should i install claude code ? how do i actually use it if i had a subscription i'm confused
How you can check that lol
You can just directly change model on the agent tab
You paid for subscription (ai pro or utrla) just for higher usage
so the google subscription not claude
Since your using antigravity... So yeah
If you subscribe on claude just go use cursor
python libary called tiktoken. to get an estimate for how many tokens a file has I think its from openAi but im not sure
I never install token tracker LOL
but what is that
Our servers are experiencing high traffic right now, please try again in a minute.
lol
I just did cuz i was wondering why my limit was gone after like 5 promts
Google ai subscription
ah k
2m+ token just 5 prompt was crazy😭
isnt claude code better then antigravity?
ik
I think bro just uploaded whole codebase
stupid file was like half a million tokens
Claude code is a CLI TOOL AND ANTIGRAVITY IS AN EDITOR
Just model issue,you can use claude model on antigravity
they are different things we can not compare them
Claude code is cli tool btw
idk i like antigrav editor but id like to use claude opus more
💀
I mean yeah but you can just use vsc with claude code cli
exactly
Pay for api key atp
but its still not a whole editor you know
api key ? claude pro ? google whatever subscription ? i'm confused
I mean token handling is diffrent
Are you new to these ai thing
🥀 🥀
looks like he is
i'm new to coding
i just want to use claude in antigravity without as much restrictions and it's just confusing
it is very expensive
The stupid file saves Images encoded with base64 in the text file
Yk what? Full 5 day of openclaw burn me 50 million token
that is-
concerning
Genius
ipynb file if you were wonding
what's the best cost efficient model rn ?
ah now i get it
For what
There like 67 model's around here
Gemini-3-flash
oh yeah, does anyone know how to show the missing convo history / open the .pb conversation file? on anitgravity
making a website let's say
Real
Cloude opus 4.6 max
funny !
Hell nah
he boutta pay his whole year's salary in one days of usage
Did someone try openclaw here
hmmm
idk
That's thing is Money burner Holy god
ye i use locally hosted model there
so what do you guys use ?
Claude Opus 4.6 Max
i have 200 usd per month subscription so its nice
Good for you have good gpu
where can i make tickets?
Cursor ultra
gpu eh kinda, m3 Ultra has a nice GPU
cloude or what?
that is also nice but i do not like Cursor's new UI
can you combine claude apikey with cursor?
Claude
Im using gemini 3 and it rush over 50 million token just few day
3 what flash or pro
whats bro doing with 50m tokens 😭
OpenClaw
Same lol
its hungry
i'm not coding full time so would a normal pro subscription be enough with claude ?
do you have it?
Open claw, ultimate money burner
Yes lol its already enough
bru its free
Based
lemme guess
gemini 3.1?
And i deleted all my open claw
lol
Like after 4 day of using it,58 million token usages
Close enough, 3 flash
what's this
ah, kinda expected it
ye
antigravity tool kit its abit nicer design and you can mange memory and stuff
it searches for everything even tools 😭
oh is it oss
gemini is a smart models, but most of the time having issue with itself.
instead of solving the main problem, it solve part of it, and do quirky thing
whats oss?
Open-Source Software
ah
i wish they made the models actually usable
they do not like following instructions
for me usually it works 75% of the time, but i must repeat the same thing over and over before it behave
" ya you know openclaw are the best ai right now"
btw, does antigravity changed the quota? it seems like I hit the limit frequently than before
Bro did not say how much it cost for one week
yeah that is kinda the issue
They always changing it
they are testing something
and i did not use claude or pro models for a week
AND IT STILL SAID I USED ALL QUOTA lmao
I think
I tried one message to opus with Credits, it costed more than 100 AI credits
i only used claude to migrate things, and ask how would it implement some feature and i just hit the limit
Let me try
🙏
I have like 944 ai credit idgaf lol I don't use for flow
hmmm i am lucky, it works longer as i love keeping convo small and telling it to output as less as possible
@misty tapir How can I open a ticket or something?
Should I use sonnet or opus
try try
OPUS
Later... I'm using it for making web game
I think same for me aswell
niceu
where can you see Ai credit?
nvm i found it
nice
whats high amount of ai credits?
what
to do what?
if you working with a codebase that needs a lot of codebase but not that complicated, give sonnet a try
but if you work with a complex codebase, its better to use opus
i just use flash for everythign
25 USD = 2500 AI credits do your math
I mean token wise
idk i never tried it but its very less like 1 message to OPUS might go to 100 Credits
k
@heavy quest i think you deleted some message
honestly i never used flash model, as im afraid of what gemini model mixed with flash would do to my codebase
trust me its scary
i get amazed by how smart ( sarcastically ) that model is
for now i trust opus only (for full access), and gemini pro (if its only touching a small codebase)
nice
i got Chatgpt Pro sub for free by OpenAI randomly
for a year i should use it
but what i've seen (from t3) is xtra high is slightly dumber than high models
hmmmm
just gonna stay on default then
t3 code feels nice to use ngl
yeah i get it
UI is a differnt thing but UX needs to be better
i have reported sm things lmao idr how much i have reported
I hope antigravity agent manager, or conversation history would be better (ui and ux wise)
ey ux needs a lot of
just open source it atp so atleast we can do something about it
as i have encountered numerous missing conversation (present on .gemini folder) after disconnecting from ssh server
yes i have had some issues like these
Lmao
Openai try to get back user from gemini
yeah but this pro plan is 200 per month lmao
5.4 is actually good tho
Should i try one month free chatgpt pro
yeah
it has SO MUCH usage literally
Gonna burst the usage
i tried today it is not ending
I mean plus
oh plus idk i think it should be fine too
If i subbed on chatgpt if my codex also going to be same?
Or its totally different (pls no)
But its pro,pro plus and utrla
If i subscribe chatgpt plus was it "pro plus" on codex
oh i have no idea let me check
Cause the pricing was totally different
We have chatgpt go here
From the pricing it should be "codex pro" sinde its same 20 usd
are you from IND
GOODBYE AI TOKEN 🗣️
i see, well try one month free i guess
it is way more than what we get for claude
from Google one Pro
GOODBYE AI TOKEN 🗣️
wdym
i dont have free plan from claude
i pay for claude api key and i cancel after 2 week
its in the Walkthough
the cost was crazy for just vibe coding some game
claude api
bro do not do that
use 200 one it gives you thousands of usd of usage
idk its a game function i didnt play that much yet since the progress was sllooowwww
0 ai token incoming
WHAT?? 80 TOKEN JUST THINKING??
@crude moon oh god bro claude opus is broking my ai credit
how i can stop it
you know what? gemini 3.1 pro better
i stopped lol
using 3.1 pro now
@crude moon how much does ai credit cost like a rate to irl usd
12 second and 80 ai credit is GONE
HOLY GOD
a dollar for 100
around that cause 25 for 2500
Good
80 cent burn on 12 seconds
i cant imagine how much it gonna cost for one day usage
not even mention opus MAX
do not
hello morning @.all
burning my pocket
i've question
who can assist me with an issue on the integratet video-creation by KI fsrom NotebookLM?
im never using gemini 3 flash
stupid ai be stupid
Ill show ya'all what i wanna tell
its the creation of a videoclip out of my sources i uploaded at NOTEBOOKLM for a campaingstart in Dungeons&Dragons
i'm playing around with tis to have i handy later when i will change set from player to GameMaster
here are some screenshots with the white boxes overlapping the szenerie which ist so important for ingame-immersion
btw do u have any solution for it?
anyone else get account suspended recently? if so, did you get your account access back or did you remain banned? I'm looking for alternatives to antigravity and the gemini suite now... any opinions which to try next? I don't think I want the subscription anymore if I'm being honest. getting banned while it the trial period doesn't do well in making me want to use more of their products...
hey
super modded lol
i hate vsx ui so i just set up myself
woah
and just put fullscreen hiding the top bar of the notification
nicee
friendly reminder, general is not the place to ask questions about products as we have #flow-general・🌊 #whisk-general・💭 #jules-general・🦑 and #opal-general・✨ where your product specific questions can be asked. Or #off-topic・🐢 to discuss other products not mentioned. Thank you!
catshire where can I make a ticket or smth like that?
Hi, to reach the moderators - DM @fringe nova , for product bugs and feedback, head on over to #1238621511401734254 . The team monitors those threads , so your questions/issues won’t get lost in the channels.
Hello, is there anywhere where i can report my Google Antigravity issue? Regarding late limit
is this a vesione kinda thing of whatsapp
what is it about?
i just had an issue with rate limits too
I wanted to ask if googler role is only available for Googlers who work on Labs, or if it is available to any googler. Because in Google Devs Discord, any Googler can get the Googler role.
is it cheaper to use gemini 3.1 through github copilot than antigravity now? my weekly limit didn't even refresh
I was on my vacation im on ultra plan and my brother is on pro plan. Somehow it says both of our rate limits are over in 1 day of using and itll refresh in 6 days. We js used it for 30mins after a week break. So im pretty sure ts wrong or a bug
why dont you jsut make a family?
like that you both have ultra plan
Wait we cud do that? Oh alr
I thought it doesnt share among family
i think if you do a family every member gets its own pool for everything expect the storage
Familys can "only" be up to 6 people and everyone has to live in the same country I think
By the way, if you have Ultra, have you tried Project Genie?
When I tried it, it was really laggy, but this could be because I'm in Europe and the servers are in the US. It shouldn't really be that laggy though. Did you have to wait before you could use it (e.g. create a world)?
That would be a perfect question to DM @fringe nova .
do I just dm the bot or what?
Yes, if you are in multiple servers with it, just make sure to pick this one. 
yeah last week ive also used it like 4 times more, reached the limit waited the 3 days
then today it ended up so quick
i sent them a mail about it
antigravity-support@google.com
but they havent responded yet
Is antigravity down for you guys?
hello
also gemini CLI seems to work even when anti gravity is out of tokens
Seems like Stitch is down as well
vesione ? could you elaborate?
Is antigravity down, it always says
Authentication Required
Please sign in.
even though I am logged it, doesn't even log me out properly when I try to log out
I get the Authentication Required message too. After a few seconds it loads with no problem.
Is anyone having this same problem where old Pro chats are just uncontinuable??
Every message i send in those get rerouted and starts new chats
I think it's now working, at least for me i can use antigravity normally now
Yo guys
Seeing people with the same +5 days token refresh issue. Does anyone actually know what’s going on? If anyone has heard back from support or found a fix, please let us know
Hello
hey google fam... what's the eta on veo 4? seedance 2 is hugging all the attention RN
Bro now its with Tokens, which you can refill
hello, I joined to look for ai-edge litert stuff but I don't know which channel is for that and this was the first server in the search results, am I in the wrong place?
hiii
hello I'm looking to create an AI for creating tiktok videos and that improves the vitality of the videos but I find myself stuck at the level of the videos it doesn't respond to but ask is it any one how to do it?
anyone know how to keep voice consistent on google flow?
can't
they don't do that
I am getting this error in Antigravity. Is anyone elase also facing same issue?
It seems that the antigravity team is changing a lot of things in the usage of model agents i think.. This seems to be causing the quota renewal process to bug 🙃
Hey everyone, I've read that Google's TOS forbids students under the age of 18 to access any and all of the APIs, which includes the Google Labs projects here. Does anyone know if there is an school or educational agreement that we can sign for students to be able to access these tools(especially Antigravity)? TLDR I'm also running an tutoring program where we explore AI tools, and I think it would be very helpful for these students to explore and play around with the goolge Labs eco system. Not sure if theres any staff members or community memebrs that could help me out. Thanks yall ! ☺️
Just use the app,as a collage student just use everything
Lmaooo we have students that are like aged 12 - 18 it's an wide range
don't want to get TOS flagged tbh
I don't see why they couldn't do it, cause if there's a "Youtube kids" version
Then again they do seem to have a bunch of requested updates on queue, and we dunno when it will happen
Also if there's a thing these Google AI models are good at, it's censorship. Nano 2 is prety much "Kids" version already, so idk what the fuss is.
<@&1366831844502274180> idk if its scam or not but it shouldn't be allowed here
hi
good day fam
hiiii
Heyyy
whats up
Its going good wbu ..?
im doing good huggie

Noiceyyyy
hiii
from
😁
haven't they lowered it?
like after they announced that the limit reset will be longer from every 5 hours to every week
it become 5 day
but i hope gemini 3.1 pro stay 5 hour.....
also funny bug lmao
High, generous quota, refreshed every five hours until weekly limit reached
HAHHH THEY NERFEF AI MODE's PRO MODEL
not first time dw
fr?
ok
obv he is tired
very tired
now i think google hasnt paid his salary
maybe it just isnt updated yet
it still tell me that cost ai credit if used
kinda impossible beside it sync with sonnet
I was kinda talking out of my ahh cause I don't know what that is, I'm assuming remaining uses, since you says reset I assume it was updating/maintenance and should be fine in a bit
like are we serious just get more customer from open ai and just instantly nerf token limit and everything?
☠️
lol
but did yours still have limits
why is gemini so buggy
i just found out that SINCE WHEN it become peter pan
nvm i open the beta
im going back to jules... for god sake
I think the limits got changed but im not sure
Hi peter pan
Yeah it's this update @finite sapphire
Back then, when you send a message, it does two phases: Thinking and Creating Layout phases, it brought accurate info and took its time.
With this new one it straight up skips both and just searches like the default model
ah I thought you ment in antigravity
What's antigravity??
we talking about antigravity
are you seriius
imagine if antigrav devs, vibecoded antigrav
Gemini cli also gets used alot
not impossible ^
I mean devs do use gemini to code
"yo gemini make a antigrav update"
T_T
I came here to see if there was a community specific to Antigravity
Anyone here got decent exp with running Cline and Antigrav agents automatically?
Antigravity patched something and all quotas went to junk, and they banned a ton of accounts.. mine included. it's time to look for alternates to all the tools I use form google. it's won't be easy, but I can't support and pay for this kind of treatment. not even a human to arbitrate with. ultimate subscription is worthless if antigravity isn't in the picture... certainly not WORTH the price.
🥰♥️
time to waste some money on a campaign to warn everyone what they are doing so they don't waste 250 bucks and then have the service yanked out from under them without so much as a refund offer or anything. it's disgusting
♥️♥️♥️♥️
... it's why I'm running cline as well. lol
Gemini mostly approves plans with its larger context.
yea, I guess it's back to cline, but I don't think it can handle the massive project.
https://vbe-link.netlify.app/
i made this
for gen z
!!!!!!!
can barely handle refactoring my 10k line PWA...
guysss pls rate this i made it
using anti gravity
needs a demo mode with no login. I ain't throwing up my deets like that.
banned?
thx for your review
how you even got banned from it lol
Probably trying to build what google considers 'a competitor' solution.
@misty relic but do u want me to add a demo in the home page? and i dont store your personnal data.
its stored oly in firebase
and not ur personnal data
not sure how it'll work. just hard to trust new platforms with credentials.
You can easily tell I'm a vibecoder... "not sure how it'll work..."
yep. got a popup, with a 1 button no context appeal with no guarantee of being read or addressed at all.
<@&1366831844502274180> is this okay if i advertise this much?
or should i delete it
okay
Hey, thank you for asking, but generally we don't allow any kind of advertising, so it had to be deleted.
so i wont
thx for informing. i was js telling about a project that i ad created but okay
💀 is that like perma ban? i never meet it send to my dms lol
Some have said they got access back in around three days; others are still waiting. I don’t know at this point. I’m not in a mood to care. I just want my money back so I can move on elsewhere.
Any sort of permanent ban with a no‑recourse, one‑strike rule is wrong, especially at these service price points and when there is no human to contact. It’s just disgusting. It shows us how the world will be soon, though.
ok, maybe some things are reasonable... but I wasn't doing naythign BAD. I was (am) building a new A2A ultra compressed deterministic logic based programming language with transparency layers and all sorts of securities. meant to be a safe zone to run all your autonomous agents in and not worry about the "what ifs" as much . pretty much dead though without the big models though. it's over 200 modules now and nearing 3000 passing test... so it's real and it does work, but it's not all wired up and flowing yet because I was in the middle of migrating from streamlit to a TUA sidecar wrapped solution for a better more OS like finish and polish then... BANNED
if anyone wants the repo to check out or fork, feel free to DM me. I dont' think I'm supposed to post any links in here. don't want to get banned 2 times in 1 week
Whwere can we report or get collective upvotes to a feature where we can ask google to change or update some things? the UI element that shows Upgrade to Google AI Ultra is so frikkkkkking annoying!
hi there! you can add to existing posts in #1238621511401734254 or create your own! feel free to use the tags to search!
not bad
It seems like they have reduced quota for all models, specially Claude
Agreed, My own antigravity models shows limit reached and I have no clue why this happen without use.
lol Gemini on my tv is interesting
Google Flow when Google Workflow walks in 
are any models showing quota yet? mine do, but my service is blocked still. you may have been quietly banned too. hope not for your sake!


Hi, I've been working on Whisk for months, and now that I'm starting with Flow, it's truly a total failure. The scenes are inconsistent; it doesn't do what the prompt says, generating too many irregularities. It's true that in other aspects it's better than Whisk, but when it comes to creating a video, Whisk is king. Unfortunately, they're going to shut it down. I was creating a very solid short film in the horror-suspense genre, and now that I've done the same scenes on the other platform, I've remade them in Flow, and it's a total failure. I've wasted a ton of credits for nothing.
I guess this transition period is going to be rough for everybody.
#LongLiveWhisk
I'll try to finish the short film; I've only been working on it for a month because I first tried to fully understand the platform. As soon as I finish it, I'll share the link so you can see that this is impossible to do in Flow, not even close. As for creating images, Flow is great; it saves you time and the hassle of writing prompts. I wish they had added those tools to Whisk.
Whoa guys did u know if u google 
Thats a bit old but yes.
what is Deep search?
anyone using antigravity in a tablet? (true question would be, if that's feasible, since, for instance, s11 ultra has a linux terminal (i guess?) )
some guy i think
🙂
Arun Mani interviewd him once
nice
Yep
Google is owned by its parent company, Alphabet Inc., which was created in 2015 to hold Google and its, "other bets". While publicly traded, founders Larry Page and Sergey Brin maintain controlling interest through a dual-class share structure, with institutional investors like Vanguard and BlackRock holding major stakes.
Capital.com
Capital.com
+4
does anyone here play with aistudio?
hey guys. how can i do ai youtube automation im just starting out. could someone please help
yes i could
ill explain
but i need to know more about ur project
not possible but i rather use cloud pc than this
so you cant use nano banana 2 on ai studio but you can on flow for free?
free account?
flow nanobanana 2 and flow usage is very very very limited
like possible 5 photo and you hit the cap
idk but for me as a pro plan its almost unlimited beside flow ai point usage
what is this bruh
yo what the hell i just made 84 images
and i can still go
i think its totally free
I don’t rlly use google search often so
Your good no worries.
I don't know when it was released initially, but it's a chill easter egg.
🦥
84 images now add one more
They said once that Pro gets 100 a day, but I think it goes up and down depending on how busy servers are.
Whats the occasion?
Happy Pi Day, people
]
hey