#general・🧺
1 messages · Page 40 of 1
gemini is just so incredibly bad its insufferable trying to do any work with it
Is the chat usually this dead or is everyone asleep?
I mean paying 200$ a month and not being able to do a micro saas or code some personal project is outraging, i don't really have a use for my 45000 credits anymore lol
Yaah gemini is so bad
It's give stupid answers
Grok is so better then gemini
Hey google improve gemini
I have a question about Google AI Studio and the subscription plans.
In Google AI Studio I get 2 options under "Upgrade to unlock more":
- Gemini API, Continue with pay per request.
- Google AI, Continue with a Google AI plan.
My question is, if I go with option 2, and take a "Google AI Pro" subscription for € 21,99 a month to get access to more requests inside Google AI Studio, will my chats be used to train the model?
I know this is not the case when using Gemini API, then my chats are not used to train the model, but is this also the case with a Google AI Pro subscription?
Since I want to use Gemini for professional use on projects that have NDA's with sensitive information regarding film projects, I need to be 100% certain my chats will NOT be used to train the models within a "Google AI Pro" subscription.
(P.S.: The producers decided not to use an API and are looking for alternatives.)
thanks for your feedback 😆 feel free to share with the Gemini team in their community, discord.gg/gemini
Bro so you opensource models like google also launched gemma 4
But you can use Kimi and nvidia and zlm models
Because it will train their model on your data at any cost
We need Google Gemini 3.1 Pro to be used as creatuive assistant regarding the finetuning of screenplays. We need the best "brain" for the job, Gemma 4 is too limited.
Hmmm
“I think you should consider hiring an AI engineer with experience in fine-tuning models like Gemma 4.”
It is is in apache licnsnse
It will finetune the model gemma 4 for your work
This is the way for your privacy
Your best bet is API request, any subscription will be used in database for ai training basically, at least in google realm
or use Gemma 4/open source models as Vansh said
It's 100 percent sure they will train their model api or personal uss
Like they are not made for privacy they want our data
I need help. Where can I reach the Support for Antigravity IDE? There is no Email adress.
Hmm you can try on this google server and google developer community
You can tell the problem
There
thx
None of my media is loading from any project is there a fix?
Where context
Give some context be specific about the problem
Tried to work on a project i was working on yesterday and now there is nothing.
anyone can help me with deciding if i should switch my agent workflow from OpenClaw to Vertex Ai
Cload is very good in prompt engineering 👌
Hello is flow having system down?
Hi I don't think so
It is working with me
when i click create, it shows nothing.
forget about it, i switch to safari and it's working
Good to hear that have fun
What is Pomelli
its literally insufferable, you ask it the most basic question ever, expecting a reply, and it goes off and does 1000000 other things.... by far the worst model I have ever used for any coding tasks. terrible, unusable trash
Why do I keep getting this
check this out for more info on Pomelli 🙂 #labs-announcements・📣 message
is whisk down?
what will Google's response be to GPT's V2 image gen
🤔
I use Google Pro, and when accessing Gemini with tokens, it always says that its busy/hot .
Also I can't even use gemini 3 or 3.1
@outer wagon any idea why?
Hi there. PS: id like to use ai in some way in my content creation scheme.
latest anti gravity update, massive improvement, just now
nice job
when anti gravity works its so fkin good
applied server tag successfully
hello everybody, I'm working on AI agents using n8n and would like to know if you guys have any tip to improve my agents, because I use just "common" agents, thank you.
oof who uses 8n8 thats so old now
me 😅 , what do you use now?
most people jsut make there own harnesses
Here in my town most people doesn't know how to build, for me using n8n is the quickly way to produce it, I really don't know how to build it by myself, can you say how?
go to your ai and ask it i wnana make a harniss from strach so it can do anything i ask it to do. have multi providers
I don't think that is just this... I already done it and my IA lead me to n8n. I think is ask too much for you some valuable information, anyway thanks.
Hi
Hello
Ultra user here. but why is this credit only 50!? i always used lower prio
hey bruh wassup
Hey
Hi 👋
Hey
Heyooo
Wassup
What you.guys are even doin
Indian 💔
Yeah what about u ??
1 prompt 50 tokens used dumb model
hi
what is this
new agentic harnish im working on
looks cool
looks impressive
im trying made it can almsot do anyhting right now for anymodel and when i say any i mean it the smallest local models as powerful as cloude 0-0
So.... new gpt image model. Thoughts? I don't like it much
Feels way more censored than Nanobanana and less good at following references
Lol
Its much better than nano
It understands references much worse. And it always defaults to a specific visual style I don't like
Hey guys. just wanted to ask if anyone remembers the dynamic view feature on gemini, it was around december 2025 / january this year but it's disappeared. does anyone know why?
AI art is real ART!
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
What about OpenClaw—can we ask questions about it?
yes
#off-topic・🐢 would probably be the best place! we try and keep channels Google labs focused here!
GM, guys
Hi
yes i can help you with it and it will work perfectly
what is your magic trick newcomer?
How do you do that?
It does not work for me eighter
I call BS lol
Yeah praps a "trusted" github tool
probably some scammer wanting login info..
yes it may not work for you but i have work on it many time
Gatekeeping it like I said. Probably waiting for kneeling down to him..
Ive been dealing with Discord scammers for around 8 years.. 😅 Prove me wrong @unreal haven 👀
To which Gemeni version does Gemma 4 compares?
Is Gemma 4 as good as Gamani 3.1 Pro for creative writing?
Teach me your ways
Hello guys
So that means we can also use it for others things like news discussions etc
Yes indeed
Oh and you know how to use
Google antigravity agent kit 2.0 in antigravity
I am confused how to use workglows agents prompts
I am confused how this thing will be done in antigravity
Very confused about it
?
Hallo guys
Got timed out on gemini so had to run on chatgpt deep research might be sitting on something nice tho
Looking into a four-stage cryostat
ANYONE else having this issue ? i have 13k credits left
accept the free offer homie
Hi there
hey! welcome!
I sent it
Any video Ideas
yo
👋
Who removed my comment
Do Gwen from SpiderVerse in royal pain from Shrek 3
is anyone else experiencing errors w antigravity??
Yes with these
I dont know where I can type about Antigravity, do you guys know when traffic is low so I don't get everytime error "Our servers are experiencing high traffic right now, please try again in a minute."?
devs are busy playing blackjack
Hi im new to all the labs feature of google, can someone explain whats antigravity tool?
Ok thank you 😄
Good Stuff
Google developer community is too reporting the same issues. Got a few members asking
Yea, I'm getting this error for like past 5 days
hello, can i ask for help with antigravity here?
the latest update made the agent auto apply all changes. before i could go nicely line by line, accepting and rejecting them. idk how to change it.
Hi we do not have an antigravity channel here. The Google Developer Community has a channel for antigravity called vibe-coders. You can find the server here. https://discord.gg/google-dev-community
You can also find the antigravity forums here. They may be helpful too. https://discuss.ai.google.dev/c/antigravity/64
like what
good morning guys
if i wanted to find help on how to use google labs for makeing a music video that has two cartoon cahracaters singing along to an mp3 i crated in suno ai .. where would i go looking?
Here's why nanobanana still wins
"photo of a mountain tanager bird on the shoulder of a man eating pasta with pesto in a room fully made of glass"
this is gpt's image
this is nanobanana's image
and..... in the first image, that bird is not a mountain tanager!!
Hello 👋
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
#general・🧺 Hello, I don't know why I can't use this application. Could you please help me find out the reason and how I can gain permission to use FLOW products?
You didn't verified your age. Please verify your age.
Tbh there should be some sort of revision or even compensation for the loss of images in this way.
They can't just really pretend nothing's going on.
They did assure repeatedly that there was not gonna be any loss of images.
Find,thanks bro,iam trying
You are welcome , as always !
Hey guys
Please help me how to use antigravity agent kit 2.0 GitHub repo their workflows agents prompts etc how this. Thing will done
the mountain tanager colorations are off in both photos but i feel you liking the nano bannana image. I think the chatgpt version is looking more realistic. The nano bannana one is reflective to perfect / clean. I think they are onto something. Cant wait to see what Google does or has cooking!
Chatgpt will do something I crazy think so because of the rise of claude
.
correctly... any loss of data.... but yeah. i would prefer more that simply restore the lost data. they have backups of backups soo cant say its impossible.
also they like to forget that many of us are paying costumers... not free users....
Hi everyone! Does anyone know how long it usually takes to get access to the Disco channel? I've been waiting for 2 weeks. I already posted my info in the required channel, but still haven't been added. Any advice? Thanks!
Please contact @fringe nova so we can help you through there 😄
disco channels?? sounds lit. 🌹
got is only good at virtual stuff, googles banana still dominates in realistic stuff
Hey, there's no specific server for Antigravity, but you can discuss it with other members in the vibe coding channel on the discord.gg/google-dev-community server.
For Antigravity support, the forum would be a better choice: https://discuss.ai.google.dev/c/antigravity/64.
Gemini has its own official Google Discord server, though: discord.gg/gemini
Can anyone tell me why 2K images generated with NanoBanana2 or Nanobanana pro has size around 8-11 MB when downloaded from gemini, but when downloaded from flow.com has reduced size (2-3 MB)???
Does Google AI Pro give more frequent quota resets in Antigravity? Some people say the quota resets every 5 hours instead of every 7 days, while others say it doesn’t change anything. Which is correct?
thanks for your question! unfortunately we don't have dedicated Antigravity channels/support in this Labs server. I'd recommend checking out their forum for more: https://discuss.ai.google.dev/c/antigravity/64
Someone on the Google Labs team needs to talk to the YouTube team. Dozens of creators are getting demonetized this week who use Veo. The two departments are saying different things. Why should anyone use Veo if there is no revenue stream through their own platform?
is this right? I have google ultra and I can't generate 1:1 nanobanana pro pics
and suddenly i have 12500 credits instead of 25k now
Ive only generated 5 pics so far
investigating this! feel free to join the convo in #flow-general・🌊, and you can file this in #1238621511401734254 for further investigating
so sorry about that! we're working on expanding specific UXR opportunities in future!
All good.
who is actually running Google Flow Music? lol im so confused now lol, music sandbox to producer ai to flow music?
please use #1238621511401734254 to submit this bug! these threads are actively monitored and your question may be seen quicker than in a general chat!
I was told not to start a new ticket because I found an existing one. No one has responded
Thank you for adding it! it can take time for the team to respond to tickets! we appreciate it can be annoying but thank you for your patience!
Well It's almost a week. I pay for the ultra plan. What ever was changed over the last weekend really made a mess. A week is a bit excessive and I'm usually very patient
hey I got the API key from google and I am a pro user, why can I not use the 3.1 pro preview?
Hello, everyOne
My name is Lee and I'm from San Angelo.
I wanna be your friend.
welcome fam
Hello
Hi everyone, I’m Ashis from India. I hope you’re all enjoying the Google Labs experiments!
Hello everyone, someone can explain why when I enter the page they tell me, try Flow
Hello. We can’t log in to our Google account. Flow will automatically renew the membership, but it does not allow us to log in. Does anyone here have an idea of what we should do?,
Visit this link: https://support.google.com/accounts
There’s a search box where you can describe your issue and follow the instructions provided. You can find solutions for login problems, account recovery, and more. Hope this helps!
Official Google Account Help Center where you can find tips and tutorials on using Google Account and other answers to frequently asked questions.
I was curious about the learn about, and learn your way tools
yeah me too its genuently infuriating
...
It's been over two weeks.
Is there really no way?
Hello everyone, we’ve been exploring the idea of AI filmmaking while putting together a small AI film festival focused on storytelling. I am curious to see how people approach narrative in this space.
Idk i wish that there would actually be like a helpcentre or a faq chat that could explain and record how many times this happens but when u look for any support u js get linked to their twitter yt and linked in
Hi, I want to be able to use Google Lens with a single touch on a regular desktop computer. I saw the Google App, but you have to open the program and click again. I just want to assign a shortcut for Google Lens, for example, F3.
it is very surprising to me how Google has arguably one of the best AI labs and software teams in the world, yet they still fallback so behind and it all looks mediocre.
for example in gemini, when I have a deep research and click copy, not only it takes a solid 30 seconds freezing up the tab, it also does it all in one line! if I want it to be like .md I have to copy it to google docs.
in generating promts it is mediocre. any tips or should i rely on claude
New models have been coming from right and left, when google gonna release new models?
i wonder what will happen to here if imagefx is discontinued and turned to flow?
I'm building an AI project for an upcoming International Expo, but my laptop only has 4GB RAM. Does anyone have tips for using Google AI Studio or Stitch to build high-fidelity demos without crashing a low-memory system?
I've made my tests. Google's image gen is still much better than GPT
try #stitch・🪡 🙂 some devs are in there an might have some tips for ya
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
Did google nerf the video generation in gemini app? i realize it outputing worse result than few months ago, it seems to be the new lite model there rather than the fast one.
Does anyone know when Veo 4 will be released?
When I want to work with Flow, it is saying "It looks like you don't have access to Flow. Check availability here and reach out to support for issues.", pls how can I correct this?
the best place for Gemini questions would be in discord.gg/Gemini !
Anyone wanna vibe code
Hi
I just did
Actually got message saying I need a peer review
how much did you pay the rothschilds for them to airdrop that perfect alias directly to your brain?
Nada neurodivergent lol
Although someones about to be paying me
Hello
Total Elapsed Time: Approximately 2 hours and 15 minutes.
• The Depth: In that window, we moved through 10+ layers of philosophical extraction, technical codification, and professional synthesis.
It couldn’t reliably relay the information back to me after.
Crazy frustrating.
The only solution is to just. Copy the chat into a new Gemini?
what is this guy talking about
Using Gemini to synthesize concepts.
Or do you mean the subject matter?
Which I realize would have been better asked in the Gemini discord. Not this one. I apologize.
Flow image maker sucks, I want Whisk back
synthesize concepts? you mean make stuff up?
AIs are good at that
hey! the best place for Gemini questions would be in discord.gg/Gemini !
I wish.
I fed it philosophical “if then” statements.
Pulling from citations ranging from Kant to half a dozen other philosophers.
Asked it if it could find literally any gaps in it that could ever make it “make stuff up”. It said “no”
Asked some control questions to test its accuracy.
Then asked it something different.
And it made stuff up.
Asked it to repeat the logic back.
It made the logic up.
I’ll refrain from further comment here. I got a reply and felt compelled to justify a response.
Or I’m just a silly goose.
What am I?
I spend anywhere up to a million years searching through messy file system trees when I can search using command line getting almost instant results.
What am I?
lol. If only.
Thank you.
Any VS code fork
I let it search for 5+ hours twice before I added it to the prompts because that specific rule you give AGENTS.md doesn’t respect that one lol
Let this be my walk of shame.
hey silly goose
BTW. @widerarrivingparties so we're on the first wave testing huh? what are we testing did Flow drop a new program?? or is this the layered "reveal" of the Gem Flash Image 3, and VEO 3.1 lite?
made forgcry useing my voice take control of my destop 0-0 kinda fire
yo whats good guys
great morning guys
I'm starting an accountability group to help reach goals. If anyone is interested send me a dm ✌️
Good morning, have a nice weekend 👋
j
what app?
Custom harniss
what?
Custom harness
i cant find
@outer wagon you should use reaction roles to setup roles for today's special
hi everyone
Hi guys, i have a bug on google flow, where the screen dies not load all my projects stay black. And when uploading a picture it gives the error: something went wrong loading your media
I have this bug too since today? anyone know how to fix it?
I have a bug on Gemini, I wanted to make an Nano Banana Pro image, I have Google AI pro subscription, but it just gives description and no image.
I was wondering if someone at Google is allowed to use veo3 in youtube documentaries to illustrate scenes?
Original and edited. Genuinely, do you think it's any good or is it over edited?
hi everyone!
Sup
Sup! Just testing the tools for a video project.
Nice, what is the project?
It's for a short promo video for a craft studio that makes traditional 'Gaucho' handmade products.
Hi everyone
I created a video but I couldn't download it
Can someone help me out please
Through Perplexity AI using nano banana pro.
No editing.
One of the best comparison tables I've seen for these two AI image generators.
https://www.perplexity.ai/search/88f51c70-778e-4803-b267-205825d7f49b
Hello
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
Anyone else's Gemini no longer making google slides? Has that moved to a separate tool or Labs tool?
Anyone know with what AI these accounts be making their videos?
Why you guys using forcecry? What are thoose?
I wanna create an AI too.
Hello! My name is thunwa (ta(u)nwa) or Kornkawin boonchom. I used to make software for golod express (https://gl.co.th) so I wanna create an OS for my dad's company (golod express), I also made a many websites like https://golod.my.canva.site, https://golod.co.site, etc.
Hi....google aistudio. is there a channel for google aistudio?
Hey we are new .. care to be friends?
good day fam
#flow-general・🌊 fighting
this is incredible
magnificent, this thing works like jules guys, but it provides everything for free
antigravity specifically
i love google man
Hi, I keep getting error saying “servers are exhausted…” when trying to use opus 4.6.
I am a google ultra user.
It’s been happening for 2 days
For me only Claude models are facing these issues.
anti gravity is free for all people i am not sure if it depends on ur subscription....
update your antigravity
I am subscribed to google ultra plan
it works fine for me, i dont know why it dosent in your side
This is really annoying. What am I paying all this money for..
Are you on a free, pro or ultra plan?
is 3.1 pro high better than claude opus 4.6 thingking
Oh
that is weird to be having a model exhausted feature
🧢
especially for a person paying $100+ usd a month
yeah, don’t know what’s happening.
thats so weird
yes that is right
i make alot of money in roblox at only 16
ok buddy..
i own games
What game? I’ve never played Roblox.
opus helps me really well, jules too
incredible
opus 4.7 released, i wonder when they are gonna add it on antirgravity
i researched and they said May
yes i do
What game?
What exactly is anti-gravity for by the way? What is it's purpose compared to other AI services?
Okay, so I switched to my Pro account and opus 4.6 is working fine. That’s strange. So it doesn’t work on the Ultra plan but works on the Pro plan…
lol…
opus 4.7
im excited for it to be added on antigravity
google really loves its users to be giving it away for free
google is generous for giving me a free trial on gemini
nah
now gemini is my full stack partner
how long was the trial?
💀
shhhh
Where are the mods.
What is antigravity?
Please we need a review on this person.
not really illegal, i invite my family members to use gemini, therefore subscribing to a fre trial
i borrow their accounts to use the model how is it bad?
alexander are u gonna come to my house and arresr me
🤣
Hey someone know how to use antigravity agent kit 2.0
it is a app like VS code but wayy better
you think it matters to a trillion dollar copany
How will they find his email?
shushushshush
Nice, this is what Perplexity says it is.
https://www.perplexity.ai/search/81b7fa2d-a252-4d0a-850a-80f1579b53cb
Oh
ok put a bounty in my head then
if my google account gets banned matter of fact ill give u 1m robux
it is worth 3000USD
ok
xDD
u are awesome get it
xddd u are pranking me
@rancid wedge
No way.
It’s not expired.
IT IS
There’s a limit to how much u can use it.
@wheat valley please confirm.
it just expired as we were speaking
So what’s he talking about?
cuz ur subscribed to a plan
I WAS then it expired thats why
IT IS EXPIRED
My plan expired so did the quota appeared
@rancid wedge
O my good just realised how old veo 3.1 is getting :(. im in to commercials and uses veo 3.1 and flow as my main tools to create content, but today i tried higgsfield cinema 3....using the latest ai rendering engines.... o my God,,,,,thats was beatiful and nice...i really hope we will see veo 4.o very....very...soon for action scenes with fast flying jets......thats bad....I really want to continue to use Veo.....
if they'd put as much effort in making the image and video models able understand what's in a reference image to keep the generation accurate as they put it in pleasing the snowflake crowd about censorship they'd be catching up with the chinese already.
This month there's been almost no update on the knowledge from the models and just lockdowns and new error messages.
hi
Hi everyone - I just joined in. Looking forward!
@narrow elk
I would ask you about something, can I DM you please?
so im new here
my pixel ten gave me some free AI stuff for 12months and told me to join here 😅 O dont even know what I can use those fore
Folks who received disco email for download, did you receive it from someone whose name starts with N and next name is a synonym of a part of a big retailer. Not giving the name in case the mail is legit
please do not ping mods unless you see someone breaking our #rules-and-guidelines・📋
Why is mariner being sunsetted when live doesn't even have half of it's features yet? Are you guys planning on dropping Astra soon?
🙂
Hey guys , I Am new to this . Feels fun to be in a non-toxic gc
I am also new bro
less go
why does this happen?
how long between when you signed the NDA and got the download?
veo 3.1
Around 10 days
what are you all building mate
damm bro nice
give it a shot and lemme know your feedback 🙂
i am also building something check on my bio
sure mate
sure thing
I am on laptop
nice
i tested your game dude its would be amazing if you add more images
yes, that's what I am working on next and here for to get access to the best image and video models 🙂
nice
you can try heist as well, that's a different gameplay but underlying both are powered by the same engine
yes, daily case is same for everyone across the globe. if you try the full investigation mode - that generates story real time based on your selection and it has more clues, evidences, multiple rounds of discussion and voting and then accusation
crazy, just add more image
full investigation has 2 images - one for crime scene and other for the evidences - but yeah will add more
like every question has its owne image
or if you could do the main source of heist and investigation should be a image like we are findings hints and stuff form image and then selection a option from multiple options
Hey !
hi
hi
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
Hello from Houston 👋
It's an odd choice creating a general channel for labs products in general, and then making specific channels for general questions about specific labs projects specifically for general questions and then specifying that general questions about specific products shouldn't be made in the general channel specifically
Is there a place I can share suggestions/ideas
Generally?
Like an idea for a lab experiment
How can I get all my old pictures I made
Hello every one i hope alll things is good am new here i wiil be happy to learnd
What is this BS?
In a previous conversation I told it to stop talking about something because it was causing it to bug out. Now anytime I even mention it, it just flat out refuses. I can't even tell it to remove it.
morning fam
Hello 👋
Hello
hey mate wassup
It's a great day I found myself in this @gloomy quiver server, my regards to all members 🙏 and I wish you to have a nice moment.
W google 🙂
Ask it to perform an internal reset and remove the block.
same bruh my pfp is trash what should i ask ai to give me a good pfp
UPDATE Phil Spencer: The False Prophet https://www.youtube.com/watch?v=ubzsByvl5h0
First video of the channel. No Kinect required.
In 60 minutes, Xbox burned its legacy and handed PlayStation the generation.
This documentary breaks down the catastrophic reveal that torched a decade of trust from the 360 era. We explore how one disastrous keyn...
Oh bro please i'm intrested in starting dropshipping
Okay man
bet
<@&1366831844502274180>
@outer wagon could you kindly look into my account as well? I sent a request to DM you but you haven’t accepted yet
dont dropship, its dead and anyone adding you offering help are going to scam you. you have ai tools man ask ai to do whatever you want
But how do you know it's a Scam?
It was a "win 1000$ a day ask me how" and I'm 99% sure the person you're answering is in on it, don't know if mods deleted the message in question or them.
I have a brother doing it
and he don't want to teach me and he's already making good money from it and he did not want to teach me that's why i will have to learn it through him
hi
hiiii
Is it normal that flow on google lags for me?
what am i supposed to do here
post your ai vid
cool. What is it
ohh
it from AI
flow labs?
Does someone know how to get grok pro
Bye!!!
less creepy 🙂
made this by using microsoft ai
Hyee devs come on why don't u share what u all are working on
when are they releasing 3.2 pro man
i don't know how AI get they Data from?
where did this "sound music art and own software come from.
bro, you csn still dropship of course, just gotta do something different. I did amazon FBA for 5 years selling private label baby products i did 5M in sales not profit, was stressful and i got basically taken down by a competititor because i was amazons choice. but dont do either of those in 2026 and brother, everything you need to know is on youtube or ask ANY ai tool?
when is google improving agentic coding in their models? currently cant be trusted for coding
hallucination is through the charts
Why I joined: I just wanted a tag is that fine?
yo bro..i sent req
guys im gonna ask a questiom, what do you think the world will be like in 3 years with ai?
Cat vids everywhere, social media full of slop and deepfakes (btw why is slowmode enabled here anyone is barely typing)
Yeah ai is going to rule the world, it already is
Is that fine tho
Ill get to you relax
The reason I joined
same
Yes, thee is no possible out ones
labs tag is cool
It isnt technicaly going to rule us
Lile ppl thought computers are gonna replace humans when computer just came out
but we actually ended up using computers for jobs
ig its gonna be the samw way with ai
Its gonna be a part of our lives
but ion think it'll rule us
Yes
RATE MY PFP OUT OF 10
I asked one day ago (it's midnight in India now)
Can anyone tell me why Stitch isn't working? It says the project couldn't be created.
maybe your subscription expired , or idk wifi is bad , mine is working fine
It must be an issue on your end , check properly once as i have no problems running stitch.
hey, welcome! we're discussing in #stitch・🪡, but TLDR; we're investigating~
Sup chat
The bonus credit points I earn every day... I can't get anything back... Why? Plz help me
whats up
I'm fine, just working on some things, and you ?
same, just went here to see if there's an error on video generation, i have trouble rendering
good, hope you find your solution :)
thanks!
hey! if you're referencing Flow, we have a few channels where you can chat with other Flow users - you can find 'em all in #server-directory・☎️ (#flow-general・🌊) 🙂
Hello guys
..? May I ask why?
hey georgia! I sent you a DM
hey! best place to speak to the team would be to DM @fringe nova or use #1238621511401734254 to submit any bugs you find or send any feedback to the team!
I’m trying to get access to disco, I signed the nda
Hi I'm new
Hey guys
welcome fam
Hi
me too
just joined in
Hello 👋
Hi Guys
Good morning everyone
Hello everyone
Good
Morning
Hello Google fans xd
hello bro , anyone participating in google solutions 2026 build with ai (h2s)...??
Guys how to download Flow app 😭 😭 😭 
i am thinking to
take part
Then?
is there any google labs tools that can make some ugc vidio for advertised some prpduct??
any of you struggle with the gemini 3.1 TTS text-to-speech program? it just keeps thinking and never produce anything for me
you don't download it, silly. it's a site
I just had the app installed 2 days ago I swear
on your phone?
Yesh
there's no official app for it, but it does have that "web app" thingy that adds a shortcut of it to your home screen
Yea yea exactly how to do Dat 🥺
idk what browser or OS you have, but you press the 3 buttons on the side of the url/link and scroll down or look for "add to home screen" or smt like that then it installs the web app
Oh yea nvm thnx
Me just dumb
i was just about to write -
👍🏻
you're so silly
Huehuehue :]
Hi
sup
What's going on
Check #flow-videos・📹 hehe
yes
Hey @narrow gyro 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!
Okay senpai sowwy
AI credit 100 credits will refresh every day or not?
Sure
just want to discuss about one of your artworks
Like which AI u use and so
<@&1366831844502274180> ^
is it me or is the server in limbo state
?
it was lol
mr beast bitcoin scammer, perhaps the automation got him.
discord’s having an outage woohoo
my friend got an email lmao
Google results in Vague AI sponsors and ChatGPT handholds me to do CRUD on Sqlite based on previous QA. Asking around this question
how does one track business metrics?
I can insert some data into DB and get piped or use grafana but that specific to developer. I want like how FAANG do .
Like I've an api which get 10M requests per day but I can't create dashboard and deliver to busienss team and also I can't handcode every api call model in the db structure and that much insertion will hungI use opentelemetry with free grafana cloud. But business team wants something like power BI and premade and also this doubt nags me, then why there's too many ETL stacks
Its a airline API so we get request in form like from, to, passengers, and some device info
Does anyone know the fix for this issue? I can not check my AI credits either. I am subscribed to the Ultra plan
Hey, I have a question about Flow Labs.
sure
So, I've been waiting two hours since my last image generation, but after waiting it STILL gave me the "You're requesting images too fast!" message.
using nano banana pro?
Yup
it was working fine beforehand
Any ideas as to what's causing this and possible fixes?
there's no exact cause lol, idk. buuut, you can try stuff like using a different browser, clearing cookies, or using VPN
I'm gonna clear cookies.
I use Microsoft Edge
Update: clearing my cookies didn't work.
I tried using Chrome, but it gave me an error upon signing in.
damn. what error
400
... that's strange
probably a BUG.
oh it's very simple. all i used was Flow. All I do is pretty much put an image into Flow and then use Nano Banana. I just write "use the same exact art style..." then whatever i want after that.
So has anyone gotten flow to work today?
whisk migration finished..... there are these blank areas in the flow projects list..... why is that? there should be projects.... and after this i scroll and scroll see the loading icon at the bottom but just black area....
Are you sure it would never load?
Flow just takes a lot insane amount of ram and badwidth to show projects, much worse if it migrated everything from Whisk to a single window
Guys do you think google will fix ag rate limits
checked in 2 browsers
i checked the one of the newer whisk projects where i generated a lot...... huge amount of lost images.....
only 10-20 migrated instead of hundreds
idk if I lost anything tbh, if you did I probably did too, gonna check better.
have to go through every project to check....
Hi 👋
hey fam
Well i might be on to something big
Interesting
I will try it today
The artwork was really unique style, style that I didn't see from a while 😅
Have fun
I really appreciate that
Hi
@marsh grove google quantum needs to get a ahold of me because of my research paper
Asserts: vacuum suitcase (zero air exposure), continuous vacuum + controlled gas only, in-situ LCVD growth, Stage XII thermal diode G_LCVD-I ≫ G_LCVD-ND, 5-7 order 4K cold-trap rejection, ³He/⁴He contrast, immediate NV readout.
Nuclear demag mK platform.
Built a digital twin that uses off the shelf components
Hi
hello
Hello
I think I need to talk to google quantum
go ahead ..
for what ?
Do you know of in situ growth
mm hmmm
I may have found a way to add mk environment to lcvd
Nuclear spin would be present
then why u need to talk to google quantum.
u can talk to any other ....
Because I dont have money to dump into this lol
then go arrange the money lol 😆
and why u think google quantum will help u??
I’m proposing a localized, vacuum-isolated LCVD process inside a cryogenic environment to enable ultra-clean, low-defect material growth for quantum systems. I’ve modeled whether pressure, thermal load, and gas transport remain stable under molecular-flow conditions.”
great job buddy, go on,maybe google can help u for that.
Yeah I went about it in how a systems architect would my uncle's one
@junior kettle one thing I found is every wire has to be anchored at cryogenic stages
noice..
Also setup with current off the shelf components
Hello
So stuff available today
helloow
okeii
Id definitely be willing to share with google lol
Well ive almost fully closed my loop and all stages should simultaneously pass
hi
hi
<@&1366831844502274180>
can anyone review my idea and prototype that ive built for gsc 2026
What you got going on?
I'll check it out, thanks for sharing.
hi, im a google fan, can you list me all official google discord servers.
Hi
I still can't access pomeli
sorry to hear this! you can let us know where you'd like to see Pomelli expand to in #1432763529697562766 🙂
Anyone else get an A / B answer from Gemini today? Is this a new thing?
interesting! I'd recommend asking in the Gemini server 🙂 discord.gg/gemini
when are we going to have a motion control function
Does Mixboard run on nano banana 2 or pro
seems to me it is nano banana 2
Kk me too trying to use it and figure out
pinging this forward here!!! Grab the role to be notified of our daily challenge!!! 
Hey guys, I need some help because I feel stuck with AI video quality.
I’m generating very realistic high-resolution images (around 2K quality), but when I animate them using Flow/Veo on the free/basic plan, the final video comes out looking soft/compressed at 720p.
The issue is:
- my original image looks sharp,
- but after video generation the quality drops a lot,
- especially facial detail, texture, and realism.
I’ve also tried:
- downloading the generated clips,
- using free AI upscalers,
- re-exporting,
- sharpening,
- etc.
But the result still looks off or “AI-ish,” especially after upscaling. Sometimes faces become waxy or unnatural.
I know TikTok recommends 1080p, and I keep seeing creators posting super clean AI videos on X/Twitter, so I’m trying to understand:
- Is the free Flow/Veo plan the main bottleneck here?
- Are people using additional tools/workflows after generation?
- What’s the best affordable workflow for keeping realistic detail and getting cleaner 1080p AI video?
- Is there a recommended export/upscale pipeline that actually works without spending hundreds of dollars monthly?
Would really appreciate advice from someone who’ve solved this problem already 🙏
Hallo
Hello, everyone
Gemini pro is the best thing ever
thank you for doing the global ping for the files transfer, almost lost my stuff!
Hi
hi
Hi
hello, everyone!
good day guys
Hello everyone!
TIME LOOP COMMAND SAYS HELLO
Firebase ai logic demo ain't working where do I report it?
https://chrome.dev/web-ai-demos/firebase-ai-logic/
yes, thats because it generates the videos at 720p, no matter what resolution your input image is. if it were to generate 2k or 4k videos, that is so much more compute which means a lot more $$. the best you can do is use something like seedance which can do up to 4k (but paid), or try an upscaler and deal with the artifacts. you can also try comfyui if you have a good enough gpu to run it yourself
I honestly really appreciate it I’ve been dealing with this major issue appreciate it so much no matter how many times that I tried to make my type realistic upscaling but comes down to my video generation
Hello everyone! 👋
Hello everyone
@cosmic urchin Hey everyone! 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!
why are you migrating to flow
whisk interface is ok, i can see where to upload image that i will use to geenrate my main visual but Flow doesn't jave that
@misty tapir @flat steeple
YO WSP
How will the new flow interface look like ?
This whisk interface looks great
Dont forget to grab the new Todays special alert role to be notified of our daily challenge!!
You can grab it under id:customizeand find the little snail!!! We'd love to see you join! 
this would be better asked in #opal-general・✨ !
hii
hi#
Are there real world applications on Google playstore or any real famous website which used firebase ai logic ?
Can I get their names or links?

Real world apps abosolutely use Firebase + AI, but here's the important reality you should understand first.
Firebase is a backend + infrastructure layer, not a visible "AI brand"
@dapper meteor Hey everyone! Just a friendly reminder to keep all conversations in the supported language 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!
Hello guys! new join here 🙂
welcome!
welcome!!
thank you!! so warm!! 🙂
Welcome! 
Are there real world applications on Google playstore or any real famous website which used firebase ai logic ?
Can I get their names or links??
Ping me if reply
I make a project with flow and Lyria, can I put my youtube channel some where to get feedback?
you can link to the project in #flow-videos・📹!
Now its day 3: Flow is not generating images. Please let us know the reason.
Hey guys, i need some people to test my app. U know where i can post a link?
I also got the same issue yesterday.
Hello
@outer wagon
hi, welcome! this is unfortunately not the Tebex server/team, so I'm unable to assist here.
hi! how can I help ya
I know this is not Tebex, but as I already told you, this person is selling leaked MLOs by just changing their names.
forwarded to #flow-videos・📹
hey is there a section to get assistance with antigravity?
haven't used it a couple months and now whenever i turn it on all i get is "agent terminated due to error".
the only solutions i see online is "disable mcp servers" but i don't have any mcp servers active
hi there 🙂 welcome. we do not have channels/assistance for Antigravity within this server (for now) - recommend checking out the official forum for reported issues and assistance. https://discuss.ai.google.dev/c/antigravity/64
yeah i looked there but there was nothing that helped. Seems just like a whole lot of people having problems with Antigravity that go unanswered.
Hi, There. Is there any updates for Pomelli??
appreciate your feedback here - passing it along to the team
hi Ramiro. [our latest update](#labs-announcements・📣 message) is that Pomelli is now available in Europe 🙂 you can grab the Pomelli role under id:customize - we ping the Pomelli role whenever we post announcements for that tool.
Is there a setting that can remove the Veo 3 and Gemini logo from my images?
These things should be optional mostly for paid accounts
I have the same issue. I have cleared cache, browser, tried going incognito, re-booted, checked that images should appear, eliminated unnecessary extensions (if any) and still keep getting message "something went wrong loading your media."
I have the same issue. I have cleared cache, browser, tried going incognito, re-booted, checked that images should appear, eliminated unnecessary extensions (if any) and still keep getting message "something went wrong loading your media."
Do other images work? or none?
It works for me today!
very good morning fam
hey there
I wish you a nice weekend 👋
we can do that
How is it done?
you can use API or object remover tool
hello
hey how can i use whisk ai
Anyone have problem with gemini cli today? or antigravity?
It has been going smoothly these few weeks then it become unusable again
still cannot generate anything on flow today 🤦♂️
mine always says this
lol i'm glad that i downloaded my images right after generating them
i knew something would like this happen
FLOW Image genaration Problem is SOLVED , I have just enable the HISTORY on and its working fine~
Our servers are experiencing high traffic right now, please try again in a minute.
i hate these messages alr
How can I achieve an image to CAD task? Is there any AI that can do this job?
Hey everyone! Just a friendly reminder to keep all conversations in the supported language 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!
yup!
hi
Whisk will be shutting down and moving to Flow AI today, if you have items that you want to download or store to flow, today will be the last day to do that 😄
More information here: #labs-announcements・📣 message
hello i generated a short hooking 3 second clip for my youtube short am i allowed to use this without any copyright problems? i have added my own cc and captions too. flow ai
This thread may help answering your question https://workspace.google.com/terms/labs-terms/
You can have a look at the FAQ page here https://labs.google/fx/faq to find more information as well
The home for AI experiments at Google
Hope this helps! 😄
is this true
Anybody notice a decline in quality with Gemini 3 Pro in Antigravity?
<@&1366831844502274180>
<@&1366831844502274180> what is this spam with multiple accounts?
Appreciate the ping! Just compromised accounts
Hello, guys
Hello!
hello
hey fellas
Google jules should be illegal for being that good to fix code without touching its a life saver for being lazy
haha appreciate that feedback!! forwarding to #jules-general・🦑 so the team can see it
pay for something that never work very bad
Sure thing thanks
Google product are best
I never regret it paying but only yt premium is the one thing i have never tried lol
klk
what was that
compromised account
oh
Good evening guys
Hello
Yep how are you
Im good 👍-hampter
AI game engines
i didnt want to test the flow app ya know id love a whisk app on ios… hahha
whats going on with that whisk would be a good app maybe people could actually make content, ya know. before the aliens get there in oUR MIndsssss
my flow account just keep spinning but never load or can i access, any tips on how to resolve?please.
Just downloaded it! Feedback for bugs should be around in #1238621511401734254 ?
you can tap the ‘Send app feedback’ on your account page to report a bug or feature you’d like to see 🙂 or tell us in #flow-app・📱, for now!
Perfect, glad to help test the app 
sorry to hear this! if this has been happening for more than a day, please report in #1238621511401734254 so our team can help troubleshoot
Is there any AI detector that truly works?
oh is that supposed to go in off topic?
sorry
yes, i really thought it was a massive update but it wasn't, let me try the suggestions
thanks by the way! still working on it!
I freaking hope so!
Looks fire right?
<@&1366831844502274180>
hi
hi
How can I benefit from this server?
hello
Well.. what do you need exactly lol
The server discusses artificial intelligence, image animation, and image creation.
well i run ads on meta for my shopify store
can it be of use staying here
?
If you are using any of Google Labs' tools, then yeah
How can I benefit from this, or what kind of help does the server offer?
I think you answered your own question earlier lol...
hi! you can learn about Google Labs by checking out #start-here・📍. from there, check out the #server-directory・☎️ to see what channels are available and what their purposes are.
.
I understand how to use it within the server.
It's not only in US...
Also i have no idea what that text says lol
"Flow is currently unavailable in your country/region."
how.can.securly.block.google.gemini
when will you all allow faces cause thats the only reason i am here
t’s such a beautiful day today, if anything frustrating happens, try to fight the frustrations and be calm." Another shares on rules: "You should always ask if a rule makes sense and if it doesn't and it doesn't hurt anyone for you to break it, then you don't have to follow it." Lessons in staying composed and questioning norms.
Hellow 👋
Hello
i just made the most coolest image of all time
what is this
oh okay
Hey guys what’s a few Google lab tools I should check out?
Hi everyone, is anyone else having issues with Google Flow right now?
When I open Flow, it stays stuck on the loading screen with the message “Create a project to start.” The project page does not load and I can’t create or open anything. I’ve tried refreshing, but it still stays the same.
Is this happening to anyone else today
Depends on what your in to @errant goblet, Flow for sure.
If you code or program, or want to learn, Jules is great as well.
Mixboard is fun for experimentation, too. I like to utilize MusicFX & then throw it in Lyria 3 as reference material. Pomelli is great for quick ad and/or experimentation for ad concepts for a focused website.
Check them out:
#jules-general・🦑
#flow-general・🌊
#mixboard・🧑🎨
#musicfx・🎵
#pomelli・🍈
There's lots of stuff here to experiment with.
Hiya 👋
reminder, general is not the place to ask questions about products as we have #flow-general・🌊 #jules-general・🦑 , #opal-general・✨ and many more channels where your product specific questions can be asked. Thank you
Hi
f
you don't have to spam same question in every channel... lol
good evening fam
is there any channel where users shares prompts?
#flow-images・🖼️ and #imagefx・📷 but it's mostly about sharing images than sharing prompts themselves
unless you're talking about different kinds of prompts
Ok thank you for info. i was hoping there it was a prompt channel where users can share prompt. Its all about prompts so
tbh they should make a channel for prompts!
are you looking for anything in particular tho?
i list all my prompt. i got allot to share it any1 need and my vault got space for more 😉
Good morning, #flow-tips-n-tricks・✏️ is the prompt channel. That’s where sharing and collaboration on prompts is encouraged. 
I'm not seeing any Disco Stuff that is what lead me here did it show up for anyone or is it not live yet?
I’ve been using Google Flow for about a year, but the past 3 months it’s been almost unusable. Most of the time I can’t generate anything—it just says the prompt violates policies, even for something as simple as an apple. It honestly makes no sense.
Hi
i am looking for help
i cannot click on accept idk why!
Hey, that usually happens when you haven’t scrolled all the way down yet. Scroll to the bottom and the accept button should unlock.
If it’s still not working, try making the window bigger or fullscreen, run it as admin, or restart the setup.
Lmk if it’s still bugging
I just got approved to Disco and got into this channel, around how long does it take for me to try on Disco for real?
Hi guyz, I wanted to get some honest perspective from people in AI/ML research.
For someone with solid project experience in AI/ML, but without published papers or formal professor-led research internships yet, what’s the most realistic path into research internships, contributor roles, or part-time research opportunities?
I’m curious whether Open Source contributions will help me with that or not. And also I'd love to contribute to any open source project here.
Would really appreciate any practical advice.
Oh okay
Nice, congrats! Usually doesn’t take long just give it a little time and you should be good to go.
<@&1173646484311384114> Just want to say, I've loving Google Labs Experiments right now. Thanks Team for launching Flow Music out of the Lab. It will go well with Flow's film making capabilities.
i had to run it as an admin
thank you it worked
Received the Disco invite and completed the form + NDA yesterday, but haven't received anything since then or a link to access the app. Don't mind waiting, just curious what I should expect in terms of timeframe.
I’m having the same issue. I got a Disco invite, completed the form, but nothing yet since yesterday. Any updates?
happy Sunday
guys there's an issue with antigravity supabase mcp, anyone else experiencing same shyt?
urget asf
urgent*
my token is set to never expire, i've been using this setup for nearly a month without any issues, suddenly the mcp stops working and i have to run sql manually
hey, someone give me a general idea what google labs does
Google labs makes new AI tools for Google
i still cannot use Disco 😭
why?
it gives you acceess of beta version of google ai
is character creation coming in veo 3
i got the invite, signed the NDA, got into this discord.
The system said i should be seeing disco sub on the left panel, but i have been waiting for a day and still cannot see it lol
is character creation coming in flow
I hope Flow will soon update to support storyboard-to-video—this is an extremely high-demand need right now.
I hope someone responds to my email
hi
Got an awesome model ive been working on
North Carolina here could be sitting on something game changing with my pre digital twin
Work in retail but went to school for computer engineering
Sorry that's just a little too personal
Nah mainly about personal details
Sup
I cant really go into details on my pre digital twin due to duel use applications
okay I wanna learn english from u and I teach u free
is that disco test only on US
i wanna be a tester but i am in SEA
yaay the ai can take screenshot of sandbox and display it in the aplication now
tbh; it looks like a vip thing 🤭 only for IYKYK crowd. [those "in the know.] i got EU regional markers atm 🤷♀️
why they don't open test at SEA region?
sus 🧐
usually there's paperwork and policy to sort through and submit for rights acquisition before any organisation is legally privileged to launch a project accross multiple global regions [excess administration nonsense that's often not worth the trouble until the product is finalised]
ohh i see
i tought SEA is populous market on south hemisphere, so i think wanna open beta test on that region
added a nicer looking table graphs
helyea! 🤘 sounds like you got something cooking! and already strategised a keen implementation of your beta launch in a region with high demand, and low supply. [just verify your data]
Ive got something cooking
its flow based programing?
?
no just a basic text graph any model can do it just how u format it visually
Anyone made anything automous agent
someone said strawberries?
?
read my name
@verbal thistle Hey everyone! Just a friendly reminder to keep all conversations in the supported language 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!
Ok boss
I may have made a legitimately unique pre digital twin that does physical error correction
Have weaknesses identified with monte carlo
Can you send me in dm pleas
does anyone know how to setup revnue cat with google ai studio
anyone know how to access the library for Imagefx and whisk? (The superior products to this Flow bs). Does Imagefx and whisk exist anymore?
The library is just gone. I have like, 2 things in Flow I had made previously. Everything else is gone. I tried going to https://labs.google/fx/tools and going to the library, and the link just takes you to Flow. Thanks!
hhhhhhhhhiiiiiiiiiiiiiiiiiiiiiiiiiiiii gugs
<@&1366831844502274180>
<@&1366831844502274180>
Hi
Hi Everyone, how can i use the ImageFX Model for generating images ? Because now it's everything on Google Flow right?
good morning guys
Good morning Sanjeeb
I made a pre digital twin with monte carlo fully integrated
guys. I have a question. Why is nobody training LLMs to 1. use timestamps when putting something in context or writing something. 2. train them on how time works
one would think timestamps would make A LOT of things much easier if they were just like, baked in into the prompt. Like, "check time. Add time to your prompt output. add a line break. Then just output a normal prompt.

