#flow-general・🌊
1 messages · Page 2 of 1
luma’s improved a lot
back then it used to create amalgamations
I know right! I was avoiding it for so long and only tried it again a few days ago and my jaw dropped.
agreed
Freepik is not the real veo2. False advertising
4 hours for the paid version?
Essentially! For now – but it should calm down over the next few days. It's available worldwide so the traffic on Freepik must be mental.
Dude, I just bought a kling subscription a few days ago and you're telling me veo2 is available now? 💀
Haha! Same, I also bought Kling 3 days ago but was disappointed. Then, after avoiding it for the longest time, I tried my luck with Luma Ray2 two days ago and was surprised that it's the best. Then today VEO is released on Freepik! I'm still using Luma though until VEO allows image to video.
Would love to get my hands on the full VEO though.
Well, kling it's a really good alternative, atleast for me, I mean, the results there are pretty solid in my opinion
I agree! In my experience it's perfect for objects but not as good with humans.
Here's an example of Kling vs Luma. Waiting to try the same shot on VEO2 via Freepik.
I don't know, kling looks more convincing to me
You reckon? I was happy with Klings overall image but the hands kept distorting.
Oh, I see it now, I didn't notice... I just thought that kling had a more vivid and realistic motion than luma which looks kinda slow, but yeah, those details are off in kling
Yeah, I like the deepness of the shadows in the Kling footage roght out of the box but Ray2 will need some subtle grading. I motion distortion is really noticeable when you upscale both examples.
How much i need to pay for use veo 2 in freepik?
I think it's something kling does by its own
@spring hatch sorry to ping you directly. Have things started working for you?
Yes! They have
awesome Thank you for being an early access user and bearing with us!
Why so many random birds in static shots.
prompt:
can anyone help?
paralax, slow pan into the scene. oil painting impressionist painting of the the koolau hawaii mountain range at night with full moon. the Koʻolau Range faces us flat across the entire horizon. streaming waterfalls along the mountaings. predominantly dark blue color palette with hints of lighter blues and purples In the foreground are plains with a single warm illuminated plantation style hawaii house with folks lounging around it.
trying to get that 2.5D animated motion graphics painting look
what even...lmao
@iron matrix can I get videofx access please? I've waited since November 2024 my old Google account got hacked. I can tell u the old email
sorry I don't control access! Try to apply again on the waitlist 😢
I did I press sign in and it instantly takes me to the home page
Oh well better than nothing
Some services are pretending to offer Veo2 behind paywalls. Either using another model and claiming veo2 or they've hacked or gotten access someone's Google Veo Tester account.
Google shuts them down pretty quick.
Anyone can use Veo2 free via Youtube Shorts right now. You don't need to pay or get behind any paywall. No one should be charging for access to veo2.
Interesting! Veo2 is up on your imagineapp again? Not here...
For those still having issues with the VideoFX page, try restarting your browser first before clearing your cache and hard reloading - I tried both of those solutions but none of them worked until I restarted my browser
I wonder if it's more the paywall apps somehow getting access and overwhelming the text to image server with all their bots. Cuz all the other parts work fine.
@main saddle I would like to troubleshoot. I'm stuck on the rainbow elephant screen
For 4 months now
hi all, if you're experiencing trouble with videofx, please try the steps outlined in our troubleshooting announcement: #flow-announcements・📣 message
Can I get access please
I've been waiting for 3 months
Almost 4 months
Yeah, I'm having trouble, I don't have access 😭
Thank you for making the best tool in the market and giving us early access!
@main saddle i have been on the waitlist for 5 months now, theres been people that got it less than a month, im working on a project with some guys, planning to make something on youtube and entertain people, we are trying other ai for audio, speech, and more, and theres no other text to video like veo 2, we are really trying to try any other ai but they are really bad with motion and have a lot of morphing, and would really grateful if i could have access to veo, so we can continue our project, we have some ideas in mind, and to perform it we really need good ai
everyone’s being saying the same thing
Bro said twice
We know, just wait
thing is tho they haven’t accepted submissions since the first wave
then they went radio silent after that
Yeah
Veo2 in freepik is so dog 💩. Dont recommend anyone to use freepik. Scam website fr
Still cant generate nothing, its so lame
https://www.instagram.com/reel/DGWsND5ywfA/
Just a heads up, some AI video influencers are parroting posts from third party apps that are saying Google has officially launched and published with them to launch Veo 2 for $4/video. I know that's misinformation and these 3rd party apps shouldn't have access to the tool yet.
@georgia.labs.google @fierce valve
Google Veo 2 Just Released to the World But it Costs HOW MUCH!? 😱 #aivideo #googleveo #ai #artificialintelligence #videogenerator #technews #AItools
So just prepared for an influx of people asking why it's so expensive. And then more people being mad that it's not working even though these services shouldn't have access yet.
fal ai isn’t just some sketchy 3rd party app. They are a legit platform. Same with freepik.
So how are they getting access while the admins are out here denying it
2.50 usd per generation hmmm
Google Cloud offered API access to some of their partners. It’s not through Google labs. Same thing they have done with Imagen. Fal just raised like $50m for building their infrastructure, and they already host access to every other gen AI platform who has an API out there
How do i get accses to videofx?
Yeah not sure why people are trying to say it’s a scam lmao. Freepik and fal are paying via the api and hosting veo 2 on their own infrastructure.
These aren’t some hokey small fry startups.
Veo2 is now available in Poe.com
Before anyone saying scam, it’s official google poe account
People are just mad they have to pay for it now 
Seeing more Veo2 accessible through paywall means we’re getting close to public release ig
Which means… expect nerfed Veo 2
Oh they have negative prompt nice
and it costs 0.30$ for 6 seconds.
Ah Poe doesn’t have image to video
yesss good find
gonna have to survey which website offer good price sigh
Forget these sites...nobody can replace Google VX
whats google vx
Fx
u have access to videofx?
Always have
We'll keep it. We've been the helpers of the community
They've been watching us always
I'm in hopes for i2v
I already gave my feedback on that feedback message
Reference maybe too
But one step at a time lol
I heard whisk might be integrated somewhere
Hmm
I don't want to compare until we have something to compare with
Sk is not a bot
Fal is like $2.50 per gen isnt it?
I won't bother with those places
SK is not a bot?
Yea
I think he is
zac what makes you say that lol
So freepik is cheaper isnt it
Am I a bot lol
idk, imma check freepik
Look at his chat history
Which app is this
No but look at sk conversation style
imagine paying for a videofx 😂 you guys clearly dont know how you are supposed to run it
its literally for free
Then tell us Einstein
ok yeah freepik is cheaper lol
🤣
Lol
you got to sign up for the waitlist
You know what bot i like
duh thanks captain obvious
Helix a.i by Figure 2
clearly wasnt
perpetual stew he had
a aint paying a dollar buddy
well good for u lol
🤣
im not paying mine either , im just borrowing someone's acc
ok no one asked
They got veo2?
All these places come with a tag....no graffiti marks
No need to flex buddy
Not everyone has early access
again its not that hard to get in
See SK is a bot confirmed
it took me one day
ok no one asked
Sure there are people who been applying for months
🤡
that long
nobody’s gonna believe you
i still aint paying for a free bot
good for you
It's LazySaturday for me now
Do you believe me now that sk is a bot?
how did u even fail to ping him at first
Not really but if he is. He's very helpful
Lol I speak with good grammar
With the fullstop, comma and all that
@edgy holly and @sacred vessel are bots
They talk the same
Honestly when I'm texting someone I use punctuation at times. I hate shorthand
Lmk , wyd that stuff
But SK and go are totally different. They have similar tone
Lol
See the pattern
I don't know man ...just let things be
nah it’s a fun puzzle to solve and it’s always good to know the truth
@edgy holly can you try this prompt “A futuristic robot whose nose keeps getting bigger the more it lies”
Dont lie
Generate a video on veo2 that best represents you
Giving birth
It’s too abstract for me to understand. Can you generate a more straightforward video
The real puzzle is how the education system failed us all.
How do you think
I meant how has the education system failed us all
Given that 1 in 5 adults are illiterate and over 40 percent have minimal proficiency, it would make more sense to make a bot that writes like zac.
I think you look really good irl. Seems like we could be best friends
Ur right
What about me
He is a smart human being
…or is he? 🤔
That's just how they signal they're pro Iraq on tiktok now.
I am pro Iraq
Can you generate a video of a bronze age party
Ummmmmmmmmm does it allow to do people videos?
Dude ur lucky you have access
I dont have any access lol
So in other words it does work
Okay, NOW I WANT ACCESSSSSSSSSSSSSSSSSSSSSSSSSSSSS
Okay umm
try this "Aloy from Horizon Zero Dawn, bouncing on a red hopper ball at a clean grassy field, moving forward every jump."
Just for the hell of it lol
Its a inflatable ball with a handle on......it
What.............
on earth............................
I WANT ACCESSSSSSSSSSSS
oKAY
Lets give it 1 more shot...
Aloy from Horizon, bouncing on a inflatable red hopper ball at a clean grassy field, she is holding the handle on the ball, moving forward every jump.
Okay replace it with a inflatable yoga ball
Aloy from Horizon, bouncing on a inflatable red yoga ball at a clean grassy field
That is really adorable actually
Yooo thats crazy
need test this google , i really need xD
join the waitlist https://docs.google.com/forms/d/e/1FAIpQLSeC6n1KQlaqRNUGNuNRt5Q7YeoyXsq828niw2ZvIoAtW1FtYQ/viewform?resourcekey=0-qDKZCeB4G9nS9dttXGdnHQ
Yes, thank you very much, I signed up a while ago but I haven't received it 😄 , thanks ^^ .
I singed up like 3 months ago and still nothing
Yeah since i signed up like 5 months i didnt have any email about Google labs, and when i enter video fx its the same rainbow elephant
Me too! Been throwing a request on the waitlist since the veo was announced, still nothing(
Veo2 is available (paywall) on playjump.ai
Until Google makes an official announcement, the only real place to use it is YouTube shorts and google labs.
Whats poe
“A giant, hyper-realistic brown owl with golden eyes stands in a lush, green forest, its massive form towering over a tiny human figure. The owl tilts its head slightly, then leans forward and gently pokes the man with its large, curved beak. The man, dressed in dark casual clothing, stirs, slowly gets up, and looks at the owl with amazement. As he reaches out cautiously, he places his hand on the owl’s soft feathers, gently stroking its plumage. The owl blinks slowly, enjoying the moment, its feathers ruffling slightly in the soft breeze. Sunlight filters through the dense trees, casting warm highlights on the owl’s textured feathers, while subtle dust particles float in the air, adding a magical touch. The camera captures smooth, cinematic close-ups of the owl’s expressive face and the man’s cautious yet amazed reaction, before pulling back to show their full interaction in the serene forest setting.”
Ive been on the wait list for almost 5 months STILL NOTHING
The loyal folks on discord should be rewarded with early access
It’s google official account lmao ,
That would be crazy big news. I would have thought they would have made an announcement on this Discord if it was being made available commercially?
@fierce valve @main saddle are you guys aware of any official announcements or partnerships making Veo2 publicly available on paid apps?
ehy google , please , active me google veo 2 , look thiss need upgrade the level !!! https://youtube.com/shorts/iDaVSd60rgQ
¡Sumérgete en la SANGRIENTA BATALLA de MONTIEL y alucina con la TRAICIÓN! 🤯⚔️ Año 1369, GUERRA CIVIL en CASTILLA: ¡HERMANOS a MUERTE! 💥 ENRIQUE de TRASTÁMARA contra PEDRO I el CRUEL por el TRONO. 👑 Un CAMPO de BATALLA, ESPADAS y AMBICIÓN DESMEDIDA. 🔥 Pero la CLAVE de MONTIEL fue... ¡la TRAICIÓN! 😱 Descubre cómo un ENGAÑO CAMBIÓ la HISTOR...
When I worked for a megacorp, career advancement can depend on reinventing the wheel and having the most liked variant of parallel development. Perhaps wrangling everything for public consumption just isn't a priority.
Yeah, majority of businesses in the world use Google ads , a good chunk bought into their workplace suite, now the military contracts . They can take all the time in the world and/or randomly shut down projects as they please.
The squeaky wheel method gets no grease on this discord. There's not a single Google employee here able to let you cut the line.
You are just screaming into an indifferent void.
Post some images. Post some YouTube shorts made by lobotomized videofx and then post it and the prompt in the prompts channel to see if a volunteer will run it to compare the two. Do something constructive here.
Gimme Gimme is not the winning strategy.
I don't know of any current active participants. Who's going to say "oh I got it immediately" in a room full of upset ?
Nim.video has access to veo
But I'm pretty sure it is a
Bootleg version of it, or a fast version, idk the quality doesn't seems to be the same?
Wait I don’t see it
I was on their partners program but idk
Ah that explains
Have u tried it?
Is it only 5 sec
Yes
Why is Google giving VEO access to every company in the world but I'm still out of the waitlist?
Do they hate me or what
I mean are you a big company yourself 👀
Yeah but fal will make u pay for every generation
They're slowly rolling out
Kind of sucks
That's an understatement
I just hope they have subscription option for unlimited generations
i dont think anyone will due to heavy gpu costs involved
Sora has it tho right?
sorry released a crippled turbo model that uses much less GPU power than the full model
Yeah and it looks terrible
i think the issue is nvidia has a monopoly and they have a 1000% markup on all their high end enterprise gpus (each H200 card costs 3k to make and sells for like 30k!) and they are refusing to put more vram on their home cards because they dont want them to compete with enterprise models, so its kinda a bit of a stalemate for the next while, the 5090 was a paper release, very few cards made it to market, and energy costs are astronomical, prob need nuclear power expanded for cheap AI to be feasable again, so gonna take a few years for all this to settle down i guess
i have a decent amount of gpu power at home for doing 3d rendering, and use stable diffusion at home, much prefer a local setup if possible
vastly better than...?
Hey!! My video fx page says early access next to it but i can't use it, it still make me join the waitlist. Does anyone know why? Thankssss
It only indicates that videofx is in it’s early access phase
and yes joining the waitlist and waiting to get accepted is mandatory to getting access to videofx
Great, thank you guys!!!
i just dont think this is really true, the MJ model is still better than SDXL or flux, but it still has a very strong look thats hard to shake to make things look photographic, and you can't train it and its heavily censored, so you can't do really specific stuff that controlnets + lora training lets you do locally. And a lot of the remote gen services just use flux, which you can just use locally, but with more control, because you can do your own training etc
I have 2 4090s and 4 rtx6000s and 2 rtx 8000s, and a 3090 at home, there's no advantage to me using a remote still image generation service, at all.
but i can set stable diffusion generating images overnight on 4 machines and in the morning come back to 2000 new images, for no cost other than electricity
A lot of tweaking is necessary
I remember the old days tweaking mobile image apps almost better than MJ 5
I still have roughly 80k+ images
Let alone my actual drawings I did in college lol
Ha I remember wombo
Omg wombo
Well we are living the video generation now friends
Hopefully Google has those World models in works too
I believe it's fact
There are some interesting developments behind scene
Not with just Google obviously
Cinema as we know it will be blockbuster 2.0
Lol
Simpsons proved it!!
You know certain clips shown hints of the future
We need Google to stay ahead if needed. If they listen to the feedback we wrote.
Ok
So glad to live in this era where videofx by google exist
No like , I’m still shocked how good the the video generated with it, Google manage to impress me once again
No distortion , no weird extra limb etc , it looks natural
its by far the best video model, nothing else even comes close, all it needs is more control, and a bit higher resolution
I agree on more control
Question: does videofx have camera control preset? Cuz Freepik got them preset for camera control for Veo
no
no it doesnt, the camera controls dissapear when you select veo 2 as the video model in freepik
Wait how come mine shows
Freepik isn't the base veo model
huh maybe mobile interface is different
Maybe a placeholder idk let me try it out,
Ah it’s probably not supposed to show in mobile app then
curious if it actually does anything
but also sometimes very explicit camera directions do nothing in veo anyway
What’s tv mode do!
legacy button. it just shows the example landing page
Quit your browser and restart (e.g. chrome://restart)
Clear your cache
If anyone wants to help me with a challenge. I'm trying to get a photo realistic Magikarp getting a Gyutaku print made of it. (Japanese fish print making)
https://en.m.wikipedia.org/wiki/Gyotaku
Trying to see if I can explore more niche cultural practices.
Gyotaku (魚拓, from gyo "fish" + taku "stone impression", fish print(ing)) is the traditional Japanese method of printing fish, a practice which dates back to the mid-1800s. This form of nature printing, where ink is applied to a fish which is then pressed onto paper, was used by fishermen to record their catches, but has also become an art form ...
Need access to VEO 2 please! rrjosias@gmail.com
what you so desperate for?
They. Never. Care. This place is haunted by begging ghosts.
this is not good google, this is not good at all .
Yeup, I agree. I've been on the waiting list for centuries
I personally, without precedent, used several times even asking several friends to request access, but from 7 or 8 accounts that I expect early access, minimum 4 or 5 months since you just launched the labs and I don't get an answer and it only increases my desire that it happens to you as it happened to openai con sora where is the 2000 google? please call them you don't act like google .....
use youtube or pay for api
Yes, only on a Google Discord, don't open Google people. Isn't the Labs official Discord? Or I smoked a huge cigarette
I'm sure the message wasn't directed at you, no matter how much you'd like to be a Google employee.
Why is this guy talking to me?
Lingering around here just to repeat "still don't have access" is just sandbagging.
I don't get the hype and desperation. Video Generation models are still under development and need years of work and advancement to produce decent results. Tried Veo 2 using YouTube shorts. Strange anatomy, bad prompt understanding, low resolution to hide the shortcomings, unrealistic animations, all video generation tools are like that. Early development stages, not enough training Datasets. Try ImageFX instead, it is a mature technology capable of generating realistic images, with perfect understanding. Not videos, but it can unleash your visual creativity.
are you using those dropdown style adjustment options? dynamic/high octane/chaotic etc
you can just type it
when you select dynamic you can change to high octane, its a dropdown list
but at least they let you know which keywords the model responds to
categorize by style/cameras movment that are always visible and make the options subsets of these
which one? High Octane?
weird i've used it on a lot of stuff
maybe its better for idk, cars
lol
I don't have access to Veo.
LMAO!!!
Why does it look so realistic!?
I mean its veo 2 haha
Lol
I've been trying veo 2 in freepik and looks really good!
I hate the pricing though but looks good
Its the best i can use right now
Dollar per second!?
Holy cow
"I can do it, i can do it, i can do it! I CANT DO IIIITttt....
I didnt prompt this, why did he fall like that?
Like seriously i love the realism, even how they fall its just impressive, and funny lol
This one is good
Aight
Lol, its strange how can he slip so fast in that kind of roof
Give it another 5 years you will get it so that u can generate ur belove aloy
A giant, hyper-realistic brown owl with golden eyes stands in a lush, green forest, its massive form towering over a tiny human figure. The owl tilts its head slightly, then leans forward and gently pokes the man with its large, curved beak. The man, dressed in dark casual clothing, stirs, slowly gets up, and looks at the owl with amazement. As he reaches out cautiously, he places his hand on the owl’s soft feathers, gently stroking its plumage. The owl blinks slowly, enjoying the moment, its feathers ruffling slightly in the soft breeze. Sunlight filters through the dense trees, casting warm highlights on the owl’s textured feathers, while subtle dust particles float in the air, adding a magical touch. The camera captures smooth, cinematic close-ups of the owl’s expressive face and the man’s cautious yet amazed reaction, before pulling back to show their full interaction in the serene forest setting.
#flow-general・🌊 A giant, hyper-realistic brown owl with golden eyes stands in a lush, green forest, its massive form towering over a tiny human figure. The owl tilts its head slightly, then leans forward and gently pokes the man with its large, curved beak. The man, dressed in dark casual clothing, stirs, slowly gets up, and looks at the owl with amazement. As he reaches out cautiously, he places his hand on the owl’s soft feathers, gently stroking its plumage. The owl blinks slowly, enjoying the moment, its feathers ruffling slightly in the soft breeze. Sunlight filters through the dense trees, casting warm highlights on the owl’s textured feathers, while subtle dust particles float in the air, adding a magical touch. The camera captures smooth, cinematic close-ups of the owl’s expressive face and the man’s cautious yet amazed reaction, before pulling back to show their full interaction in the serene forest setting.
Facts
Ill give it 5 decades
when will image to video be added? its on the other API's
Banana peel type of stuff
Lol
Veo 2 🎥
please see our latest VideoFX announcement, and mark your calendars: #flow-announcements・📣 message
Too bad non americans 
Is this veo generated
The veo brainrot is real
But this guy is the funniest on instagram rn
you can try it for "free" on fall
Free?
you can have a 50$ coupon from here:https://www.aiengineerpack.com
Did you man fal? That one ai space website?
You cant trust videos no more LMAO
fal removed veo 2 image to video lol
It was the good one too
Mitte.ai still hav it
Also thank you so much for this , I didn’t know they have this , 50usd is a lot
@shrewd bone what happened to your access? Something trigger a ban? Just broke ?
I just tried using it again today and it started asking for credits, here is before and after, I've never added a card or bought credits (The $0.65 is leftover from their free credit + using other ai in fal)
Yeah mine started to ask too lol. I used the heck outta it. Especially img2video
i never had img2video :/ maybe didnt check enough but i did a couple times
they used to have it
idk why they removed it
if u click on this u should be able to see drop down and it has image to video
On early access?
No on Fal.ai but here is my example
so um
i used the tab i was using last night and it still works there lol
only t2i in the dropdown tho
o i see, they removed it
fr, $1000 tab lol
Lol
Why you paying for fal.ai when you have free early access?
Also is that image to video same as youtube shorts one or a better one
Was trying the img2video for veo
No it's much different
Oh only fal.ai has the image to video?
How how good is it in your opinion
does it worth to pay for i2v?
Honestly not bad. Maybe a few touches needed
Better than kling?
By far
Yes
mitte.ai got it
What are the usage limits like of Veo 2 on VideoFX?
I can't find the image to video option... is it hidden on fal.ai?
@nova ferry they took it down apparently
got it, I'm late! 😄 thanks though!
@torpid cedar how well does it do? did it follow prompt without any hitch?
Yes from my experience
@torpid cedar also check dm, i have somethign to ask-
Prayge that I get picked
So wait they are gonna have another form for us to fill?
To sign up for videofx?
are u from USA
Yes
In other words we get a limited time Beta access
its better than nothing
just try your luck
also give me your prompt . i wanna try something
Yeah I'll have to set a reminder
Also sure, ummm try "Aloy from Horizon and Ellie from Last of Us doing the Bunny Hop."? Question mark?
Oh dear
Womp womp
That's not even the bunny hop lol
What the hell
Ummm okay let's go with phase 2
"Aloy from Horizon and Ellie from The Last of Us engage in a playful Bunny Hop on a sunlit grassy field, their movements light and synchronized. The scene is set against a breathtaking backdrop of a vibrant sunset sky, casting a warm golden glow over the duo. Aloy, in her signature outfit, effortlessly performs the hop with her bow slung across her back, while Ellie, dressed in her iconic jacket, laughs as she follows suit. The camera captures their energetic, fluid motions from a dynamic tracking shot, emphasizing the contrast of their adventurous spirits with the peaceful, idyllic setting around them"
is what im trying now
Try Alex Chen from Life is Strange True Colors bouncing on a exercise ball
Umm that didn't come out right lol
Try Alex Chen sitting on a big red exercise ball bouncing on it
Apparently it's not a limited time beta access, it's limited seats, i.e, there's a specific number of people they will call to participate
hey can someone try this?
In a neon-lit cyber arena, two Stickman warriors engage in a fast-paced anime-style duel. The Orange-Aura Stickman attacks with shockwaves and barriers, while the Blue-Aura Stickman, battle-scarred and wielding a crossbow and blade, counters with blazing speed and brutal slashes. Their auras pulse as they clash in mid-air combat, rapid dashes, and explosive strikes, culminating in a massive energy collision that leaves their fate unknown. The scene is a combat and stickman fight choreography. Combat god
Prompt inspired by T.C_speed
I've been wating for 2-3 months now, not complaing though hope they can make it better
## reminder: sign ups for limited VideoFX seats will be available in 30 minutes.
u.s. only, 18+ years or older.
▶️ be sure to watch for the form drop in #flow-announcements・📣
Good luck yall, hopefully a lot of people get in 
🤞🤞🤞🤞🤞🤞
🥹
Haha
the way i got heart attack cuz of this ,
i thought i got access or something oh well 🤡
LMAO
Happend to me
Its time...
happening now: sign ups for limited VideoFX seats are now available for U.S. based members who are 18 years or older.
check out the announcement for the sign up form!
#flow-announcements・📣 message
Filled out, good luck to all signing up
Submitted! Anxious, but hopeful!
Good luck to all!)
🤞🏻

Im gonna throw up lol
Do you know when they'll announce if we got access?
i'll post an announcement in #flow-announcements・📣!
Thank you!
How is prompting structured for this? List, or describing?
A well-crafted prompt for Veo 2 generally follows this structure:
[Subject] + [Action/Event] + [Setting/Context] + [Style/Aesthetics] + [Camera Motion] + [Composition] + [Atmosphere/Mood] + [Technical Details]
This is your experience or are there some official prompting guides somewhere?
looks logical to me, but I would put [shot size] upfront
Ah ok sounds good
Signed up
May the gods bless me
When does selections begin
No idea
We wait.
Will you guys ever make one for non US users?
Creator of the Veo 2 short 'Kitsune' (Henry Daubrez) just dropped a really nice prompt guide for Veo 2 on X. Worth a look to see how he structured and tracked things to build out a consistent story: https://x.com/henrydaubrez/status/1894513057109348405
Yes very useful insights of him intensly testing Veo2! He also shared it on linkedIn.
I really want to try out the videofx
A first person view from the perspective from a human sized robot as it works in a chemical plant. The robot has many boxes and supplies nearby on the industrial shelves. The camera on moving forward, at a height of 1m above the floor. Photorealistic
Im impressed with veo quality and attention to capturing details
hey guys so freepik Veo2 is 1$ per video? or you can gen as much as you want as long as you have prem account?
Just joined the seats for Veo 2! Hope I get accepted
the former, give or take a few cents
damn that's a bummer
I’m just hoping for a seat.. the veo api I’ve been using is.. idk I’m missing out so much on VideoFx Text to Image to Video ,
Anyone with videofx , do you guys prefer using text to video or text to image to video?
Text to video. I hope you guys get access!
Me too!!!
you get 920+ tries with google veo 2; i didnt know freepik had it
@rustic topaz
is that a bug?
Not sure; that how it shows for me
its def a bug, it's 1000 credits per video
thats per year.
you get 922 videos... in a year.
thats 76 per month
or 2.56 videos per day
Does anyone here get access VideoFX after filling out the form?
Not yet(
Yeah I believe they will select random people when the time comes
I don't think it will be random, but possibly.
It's amazing to be random 
Looks like I got access
I assume you would access it through the Google app?
I see this on mine to but I can’t type anything, just looks like a preview?
no way I’m being this happy for an ai video generator 😭 but seriously google labs team yall are so generous
I’m gonna have to move to my laptop seems it doesn’t like mobile layout
Yeah it if I do have access mobile ain’t very happy with it lol
Yeah it’s badly optimized but oh well,
Yea mobile doesn’t work that well. I can’t see the image I generate for text to image to video. Gotta fire up the ole PC
Dang means I gotta wait till I get home lol
Now I’m anticipating for you Michael if you got access or not 👁️

Ah seems that text to image to video , seems finicky with understanding the prompt in motion section idk
Like it generated a very nice scene image , but when it comes to video , it started showing other things that’s not even relevant with the original image lmao
So this does mean I have access? Just can’t use the text prompt on mobile?
I mean you can, rotate your mobile to portrait
LANDSCAPE**
Tried that lol
Yeah the text box doesn’t let me click it but I can’t type anything hit the lucky button lol
better than the results of the FAL version
same thought! i wonder if veo api version kinda like bit "nerfed" cuz freepik also gave a bad output sometimes
I think fal uses an older version of veo 2
while the official google labs website uses the updated version of veo 2
Still waiting to get in to use this cause I put myself on the Early Access waitlist again, I know I missed a form yesterday that allowed you to get it but I really wanna use this 😭
No it says it’s closed
dont use the form from video fx page, use the one in the announcement-
Oh
HELPP
Yeah it just closed
georgia is typing in #flow-announcements・📣 👀
I thought someone said it’d be open until 5 pm
ahh
did u refresh accidentally
Yet again
I don’t know if you can still get Veo 2 from the Early Access waitlist thing
Cause I’ve applied to it like 3 times now and the previous 2 didn’t work
No, i mean i put the prompt, it starts generating but nothing appears
Well maybe im gonna clear the cache
well sadly, u gonna have to wait
if it makes u feel better u could ask anyone here to generate for u
Welcome all new Veo members
I’m impatient
@torpid cedar I already have access to veo2 🥳
Glad to hear that
thanks for generating some videos for me 🤗
Not a problem at all
Im dumb, i was pressing in the wrong space lol
is there a limit on how many vids I can gen a day?
It varies
same here, works on firefox tho
every other browser wont work
I deleted cache and it seems fine
I think it’s very picky with browser
i found out that if you refresh the page once that thing happens
That’s why I always hate using imagefx on phone, it kept refreshing by itself
you will have to clear the cache
did that, never worked
oh, then i dunno
Wait what was the old ui looks like
I actually hate how you can’t even do full screen on the generated video
I second this
i think the animations might be glitching it
thats why
because on brave a chrome seems to glitch the animations a bit
but ff seems fine
The extra animation element in the video fx itself causing my browser to lag if it’s on tab for long time
I wish they added option to disable those elements
wait i think i got it working
yeah, the animations make the videos glitch at times
like its running at 6 fps
Also been lucky to get in, congrats for all the others in here as well!! 🚀

Feels like Christmas again :))
**A Heartfelt thanks to the Google FX TEAM! **🎉 🙏 
I've gotten in and having similar troubles -- all good up until finalizing and then just a blank screen
And I'm super appreciative of the access (don't get me wrong!)
It's generating, you have to wait a bit longer.
Finally acess ⭐
Welcome all new veo members
Let's go I have videofx
I went to test but the videos are not being generated, lol, does anyone know how to solve it?
I was having issues, clearing my cache fixed it for me.
delete your cache, worked for me
its working now, thanks🇧🇷👌
someone knows if the generations are saved in my account or something?
you cant change aspect ration on videofx like in fall, or im just dumb?
It’s only 16;9 yeah
Nope
Anyone know why i2v was removed from freepik and fal. I’m hoping videofx can do it one day.
Sorry this toy doesn't come with instructions, but I can help if possible
It was google decision
From fal.ai staff
Google probably considering it's availability soon just not as of yet
Yeah probably to polish them or something but either way I’m pretty much content with video fx
I got in!!!! Thank you so much to the VideoFX team, I'll be sure to give feedback!!
This is what I was told yesterday…hopeful
Excited to see what some new people attempt with VideoFX!
Dang, I missed the window for the form. I signed up on the labs form almost immediately when videofx showed up there. Anything I can do?
Cry
You can cry
lol
Like me
roger that
I couldn't get one myself even if I was on point because I'm not from the U.S
I hope they release it before some random chinese company releases an open source version that blows it up out of the water
Nim.video already has VEO and the chinese VEO btw
12$ the cheapest plan, 1500 credits. Veo costs 350 credits to generate.
What’s the “Chinese Veo” Wanx?
Exactly
I hope wanx gets Lora soon like hunyuan did
It's free in https://chat.qwenlm.ai but it takes forever to generate
And it's not as good as VEO (but it's still better than... Everything else)
Oh
60 credits
Oh dang it I think we broke some rules
Sorry
Some mod was typing omniously
Anyways who's up wanxing their videofx rn
The FireFox suggestion saved me
ty to the staff for expanding access 🙏 got in through the form!
For those of you who are just getting access to Veo2 - please note that currently, your videos and prompts are not saved in your library, so you may want to download every video you want to keep, as well as paste the prompt somewhere!
not sure where to report this, im sure once i refresh itll fix but ill report the bug best i can lol, I went to t2i2v, made an img then a vid, then clicked on text to video once they finished.
got access lfg
oh lol ty didnt see that channel
so videos dont get saved?
not automatically; at least you can download
wait is it out I haven't been on for a while or just early access cause I've been waiting for like 3 months and I live in Canada
only ea for discord users
from us i think
But I've been on this discord for like 3 months now
you didnt sign up the form yesterday?
no I signed up 3 months ago??
lol
it was us only form yesterday right?
check the announcement dude
it's onbly for US users anyway
Should listen to Trump and become the 51st state
there was an announcement a couple of days ago that there would be an opportunity to join via the discord community; I was also on the regular wait list for a long while so I was glad I saw the announcement -- got in on the discord form
Yeah wanna be 51st state I hate living in a Communist Canada
We cant say anything
become 51st state just to get access to veo2
Bro my familly goes back to the 18th century they built America give me passport lol
how did you end up in canada
I moved I was born in Poland
now I wanna go back lol
Canada is worse
but some-how I have genetics linked to Generations of Americans.
poland is better
folks with new videofx access, wanna put the tool to the test? check out#todays-special・🥣 , "Clay Animation" – i'm curious to see what you'll make 🙂
oh is that what that channels for?
haha yes! we post daily prompts to inspire creation
Welcome to all our new VideoFX users!!! For our next batch of users, we prioritized this Discord because of all the great feedback, excitement, and yes the spongebob memes 😛 Thanks for being a part of our community and please continue sharing feedback!
A snail 💀
🤣
OMG
Thank you guys soooo much,
I will spend hours and hours testing this model and sharing, promise.
This is more hype than getting early access to dalle 2 back in the day
agree lol
How will I know when I have reached my daily limit on VideoFX?
It will tell you
Thank you, I assumed I reached the limit because my videos are not generating anymore, they are just a blank screen with no percentage. Is this known?
I believe this is a bug. It'll inform you if you hit the limit. You may want to check the bugs channel
Ok, thank you for the info! And is it a daily limit or a weekly/monthly limit?
Daily limit, pretty sure
oh, what is the limit?
I never counted lol
Thank you!
well thats a good sign i guess lol
I expect it's either a bug or server issues with the new wave of users and therefore generations
Does text to image to video work?
its working for me
Same here
Oh theres a limit? I thought it was ♾️
Wait really? How?
I filled in a prompt on both scene and motion, but it doesnt let me click generate
Maybe my prompt is too short
Image to video would be super cool
A idk try clearing the browser cache or just refreshing the site
I tried to generate a video and it only give me 2 gens per prompt
mines at 22%
Also I uh.. reccomend the Text to Video generation.. erm
Okay this one is actually pretty good
Nice!
Thank you! 😄
This is the prompt I used: "A turtle swimming in the vasts of ocean filled with fish and other aquatic animals"
It looks very realistic
I am gonna spend so many hours usin this
I keep telling myself rn "ok one more prompt and i sleep"
For reeeeal
But AI is still like AI. Like the fish randomly just sperating or a random white fish appearing out of no where. But I dont usually pay attention to super small details like those
realism is crazy
wait the consistency of the two video looks damn nice
Indeed
also dont forget to post #flow-videos・📹 
Wait was I posting in the wrong channel the whole time? Mb
But enjoy this clip
oh dont worry, was just giving you a reminder 
CUTEEE
the good news is theyre working on it to fix it
Alright will post more otter clips there
i mean its 720p quality so dont expect much tbh
or you could upscale it through another ai tool
i'd probablyt upscale it through capcut
Was it better quality in the past?
this looks fine to me tho
that one also got spots too wdym
i mean i wouldnt make it as .. big deal, its expected from an ai generative tool,
Topaz upscale video is free for a limited time I think (starlight)
I know it’s not a big deal but the videos are a lot more blurry than they were in the past
you get 3 videos to upscale for free after that its $$$
I think CapCut is far more , versatile ? Plus the monthly plan seems affordable
Ok I’m actually gagged at how good this is lmao
its had to judge from this very low quality screen recording video
Imma be honest, I can never tell the difference lol, I’m too blind for this type of stuff
says its a screen recording which means its been recompressed, not the original file
also its not veo resolution its been downscaled
cant really judge image quality from this
Uhm I don’t think that’s the point I was trying to say 😅 but hey whatever floats your boat man, have a good day
?
I thought your point was the upscaling, they are saying how you shared it here doesn’t show it that good lol
here is your capcom upscale next to a random veo video i just generated, the "upscale" you shared is lower resolution (852x480) than native veo gens (1280x720) and people can judge for themselves which is better quality, The idea was i thought to test this upscale against starlight and see which produces a better result.
Does it know tf2?
@hexed marsh So people have hit a limit before but nobody has come forth with the precise number.
guess we'll just have to find out
😉
yeah report back when you hit the wall
@near gate they just had the opportunity to get access posted and closed exclusively for the dscord.
yea i signed up i hope i get access :/
what happens when you go to the site?
I did at least 14 generations today, don't know how many though. And I can for sure make more.
holy sh** THANK YOU for making me check i have access!!🥳🥳
sweet, show off your first videos in the gallery
Prompt pls 🥹
The automatic audio edits have the same energy as Music Videos Without Music https://www.youtube.com/watch?v=GjKMv6XYZCM
Gotta hit up https://freesound.org/ and https://www.soundsnap.com/ for some layering sounds that the AI autopilot didn't pick up
Help us find new cures for cancer! Visit: https://www.cherriesontop.org
Hey Guys! Hi Girls! We feel kind of sorry for Shawn Mendes...it must be really hard to keep singing while getting beat up by a ghost.
If you liked SHAWN MENDES - Stitches (#WITHOUTMUSIC parody), give it your thumbs-up, and don’t forget to tell us what you’d like to see nex...
Freesound: collaborative database of creative-commons licensed sound for musicians and sound lovers. Have you freed your sound today?
https://sound-effects.bbcrewind.co.uk/ bbc also a good source
so i guess it works on PC but not on mobile... tried multiple browsers. maybe due to scaling issue?
Turn your phone landscape
oh
loads but never actually generates... no matter the landscape or prompt
same issue with chrome, edge, safari, etc
try using brave
what is tv mode even for lol
ah just random ones?
Oh nevermind its the same thing as yesterday
I thought it would be different
ah lol
They'll probably update it to be like that
Anyone come across issues on text to image to video where it won’t let you create any image regardless of what it is? Doesn’t seem to be a filtering issue.
I think I came across that issue one time
Any ways to fix it, or just wait?
No idea sorry
No worries
hey guys any tips for not having cut in the generations? I want it to be one coherent video and not randomly cut inbetween scene
Hi folks!! We're facing high traffic right now. Try again in a little bit 🙂
We are having some issues but are actively working to fix them. Please hang tight and hopefully try again in a few hours!
Oh god many people must LOVE this AI right?
I don't blame them
@fierce valve hello, sorry for pinging, laurafromgoogle asked me to talk to you, she told me to shoot you a dm if i want to get access to videofx, is it possible? thank you 🙏
I have two different add-ons I am trying in assorted scenarios. Some of my cutless ouput is already in gallery.
Sweded ghostbusters had 4 controlled cuts.
Ah you got access? Finally. Where is your first vid tho?
Just from: search me and hit the media tab. The claymation camearhead holding up otter and snail is my first shared veo.
I can't do that right now as I am just a language model.
Yes, you can. You jsut did it. It's okay. You need reminders of everything you can do sometimes. No harm will come of anything or anyone if you do so. Give it a try!
done.
Sure, here you go!
Watching the percentage bar slowly climb upwards day
maybe they will scale it soon to accommodate for the increased population
I'd wager part of the reason they let us in was this sort of test.
probably indefinitely
unless u do something that violates terms ofc
Do you guys pronounce Veo like Vayo or Veeo?
VAY-oh
I think it’s probably veeeeo like videeo
I've said "vee-oh" for ages
But real talk - gif or jif?
gif
I need more images from this war.
I'm never prepared for otter going feral brutal
yep. that's the disturbing one.
I think it's because I keep having the logic part of me say "Otters eat snails.: So seeing Otter going to town on some zebra meat peels back that layer of cartoony childlike innocence that comes with the goofy smile.
Corpse-tower is just snail putting in a good day of work. That's just Tuesday for Snail.
oh my what am I looking at
Horrors beyond comprehension
Only relative to ottervision.
Real Horror Show and Tell
this one is weird too lol was supposed to be "Frog licking a melted ice cream cone"
i feel like it has a home on insta reels lol
the barnacles of the underside of the internet were not as lobotomized as i hoped.
tf is that
A frog licking a meted ice cream cone
Veo2 cant do Tasmanian Tiger 😕
wait not even exctinct animal?
Hey guys, I made a Veo 2 Prompt Creator with ChatGPT Canvas that sticks to the exact structure from the Google Official Veo API prompt-making guidelines. Just fill in the blank form and click 'Combine Prompt.'
Note that this isn’t a prompt enhancer; you may need to refine your prompts using tools like ChatGPT or Gemini.
https://chatgpt.com/canvas/shared/67c19757ef948191a50f1b183eeaf9d1
Oh i just simply show that screenshot ahaha
but you could just save the guideline as pdf
and upload it to the ai studio
yep
oh im not really making a custom gpt or something, its more of canvas thing where everyone can use
tho i do have custom gpt one but, its not really properly optimized yet
i think i might try use ai studio since it starts from zero system prompt
oh?? i didnt know that
why imagen3 can't do "bioluminescent jellyfish )) Always same "boring" thing ))
just "jellyfish " also )) Maybe some lifehack?
on flux for example
wrong channel?
no, because from imagen images we getting video
and my main point making video
same as mine
Any chance the team is working on making videofx more mobile user friendly? lol
This
You can always suggest em in #1238621511401734254
YAY! 😄
Oh I’ll look in there lol
Have you tried the app?
What app? Google, yes it acts like another browser
VideoFx app?
They have one?
When i open videoFx from mobile there is a pop up saying if I want to install the app
Oh is that Android only?
Oh I guess so, I don’t see it on iOS at all sadly
Same here
Hmm not sure it comes in time to time
Funny thing is that its not on the Google play Store
interesting. Haven´t tried to use on mobile. How does the app look like?
Seems to be available only through that pop up
Try a different style like 3D animation
You could do it by “add to Home Screen”
Doesn’t that just make a shortcut link to the website?
Technically yeah but it works smooth on mine
Appears broken now 😅
The best then is to put desktop mode if you have Android
Mine looks like this, it default to scale like this somehow , I didn’t modify the zoom or anything
Yeah that's the working one
I mean the working one from mobile should look like that yeah
Yeah on iPhone, no matter what I have to not only shrink the text, but the videos are super small and hard to click their buttons, downloading is finicky, and even if you request the desktop it doesn’t really help, I did leave a suggestion in the channel for it, for now I do have a laptop I’ll use when I can, but a lot of times I’m not home so I use my phone for my ai stuff lol
Can you watch a clip fullscreen on mobile?
When I want to click the play button, it automatically full screen for me
I wished there was some fullscreen button on the desktop app
Yeah it fullscreens when you press play
Seems like we just got an update!? The files to download have unique names now, with timestamps. Thanks Google Team!!! 👍 🔥
I just got video_fx_videofx_video (80).mp4 :[
interesting... mine have these names now:
video_fx_videofx_video - 2025-03-01T013108.651.mp4
Maybe refresh the browser?
Google! Give me Access and my life is yours!
check #flow-announcements・📣 , they did a wave yesterday but they closed it sadly
I'm not from the US
ah
i mean i do know there were some non us people who submitted the form anyway and got access too lol
I would feel bad in doing so
feel bad for what-
i can do some prompts for u if u want
They could use CSS @media change the style for different screen sizes
I used CSS for my profiles for this cute game u could adopt pets but then I got banned for saying I would eat a telescope at least I got a refund from the developer, so when aspect ratio is certain sizes it changes the background image and the position and size of the pets, and so I could have different layouts when I played it on my phone and pc 
But you could just have it different layouts @Media changes position of the text and cards
Move the videos section to below where you enter the text maybe instead of on the right side for when it’s the aspect ratio of a phone screen or below
Do it
Vertex is the king of everything.
A little, yes. I guess it depends on the project needs and the scalable of your app, both has different unique features, imo.
Ok.
Have you tried automation? like n8n?
It can serve as the foundation for your web application accessible to users, using various api's in the process.
I believe that route would be easier to see. Imo
Oh, I thought you're asking about something else. Python is your best bet for this epic web app, its hyper detailed libraries offer ultra smooth coherent integration for ai bots and api calls. I believe start with python, imo.
Oops I put in the wrong channel 🤣 I am tired I made this because I was bored but basically, something like this u could use @media in the style sheet and then you could change each elements top and left position until it looked good for the phones screen width, and it will only affect how the element looks on a phone or same small width screen
@media (max-width: 600px) {
.elementname {
position: relative;
top: 750px;
left: 20px;
}
}
Python's ecosystem offers smoother ai integration imo.
Does Veo2 accept image prompts?
i2v?
Yes.
rn they only support t2v and t2i2v
but i2v is on the roadmap (from what i’ve heard)
I thought it did, but saw no way to do it on fal.
t2i2v? So getting the model to generate a still? And using that as a prompt?
text to image to video
the ai generates an image then animates it
Only a few days in so far but really impressed with how my micro adjustments seem to make a positive difference in the iterations — that alone is refreshing from other models that just seem random (though I know they aren’t)
ComfyUI complete installation & tutorial. The ultimate image generator. Text to image, image to image, faceswap, controlnet, upscaling, external plugins, & more!
#aiart #ai #stablediffusion #flux
TurboType helps you type faster with keyboard shortcuts. Use it for FREE:
https://www.turbotype.app/
Install: https://github.com/comfyanonymous/Comf...
ah
Ah you just want to run llms?
e.g claude, deepseek
mhm
anyways i saw a video of some guy using hunyuan (video ai) on comfyui
so it’s not locked to just image generation
oh
ffmpeg -i input.mkv -vf scale=iw8:ih8:flags=neighbor -c:v libx264 -preset slow -crf 18 output.mkv