None of them actually work yet reliably. I feel like Audio gen is miles ahead of video as far as consistency goes. Music hallucinations are there, but you get multiple takes to choose from and it rarely affects both. Videos come out with wild mutations, jump cuts to unrelated scenes, or they don’t animate at all and they burn credits. Even Kling has a large number of generations where I could only use the first 2 seconds out of 5. Luma is like 1-2 out of every 10 that are usable. Maybe Runway is better idk yet.
Technically speaking that's not a zoom out. If that was done with a physical camera it would be a dolly out. The difference is that in an actual zoom out the relative positions of objects doesn't change.
They also got Pan and Tilt backwards (Pan = side to side , Tilt = up and down). "Horizontal" and "Vertical" have proper names as well (dolly and boom/crane/pedestal, respectively) but I get why they went with simplified terms for those.
After some testing with both paid Runway Gen-3 and Kling AI, I can confirm Kling AI is THE best video generator. It took an image of a voluptuous girl I created with SD about a year ago and it brought her to life for 5 seconds. It was amazing. I'm definitely going to be getting the subscription so I can continue the video up to 3 minutes!
Dont forget this is a 1 week half price discount, you have to count with the crossed out prices. And the Runway $95 tier has unlimited slow generations when all credits are utilized.
@@WehrtEuchEndlich really good point. That adds a lot of complexity to the comparison actually. Tbh I didn’t even know for kling that it was a week 1 discount. I thought it was a marketing tactic like temu 😭😭
@@roofoofighter Maybe Matt was right, it's just a tactic to create FOMO but it's never intended to be the supposed normal price. Still don't think it can compared to Gen-3 Unlimited if you're doing a ton of generating.
You missed the image to video to image end point . Good video I really wanted to see that feature if it was anything like Luna image to video to end point
Kling is good if you want realism. If you want to do cartoon or drawing styles, Dream Machine by Luna AI is better. I hope RunwayML gives image to video with Gen 3 soon.
I feel like Kling is specifaly especialy trained on the stuff they showed in their promos to cheat a little bit Everytime I upload a photo with some sort of piano (even a closed e piano), it tends to just add more of the acoustic stand up pianos shown in the promos as soon as the camera moves a bit to the left or right a bit. Also Kling is way better at generating pandas, than monkeys for example
They need to add end frame feature just like for what luma dream machine AI has. Kling ai and runway 3 definitely need to add this cool feature in their next update hopefully.
I wish there was camera control on image to video, they could easily animate things if an image is given, like Leonardo AI. But despite more information content from the image, why isn't camera available? Also there is an extremely great need for higher quality, and resolution selection on image to video. But overall I would prefer Kling over Gen 3 because Gen 3 is supremely expensive and inefficient.
How come Genmo AI have default almost 2K high quality videos but other AIs do not? Even if Kling and others do add high quality options, they would make them extremely costly now. Wish they had default 2K quality like Genmo AI does.
Matt is there a way for us to download the source to any of these AI to video, also which AI to Video generators allow you to keep generating one after another of the same project? Someone needs to start an open source free AI to video Editing Software for desktop like davinci resolve.
I still think Gen-3 is the best option for serious creators if only for its unlimited plan. Making anything presentable with these tools takes a lot of generating and tinkering with prompts so you are going to burn through a ton of generations trying to put together something quality with a cohesive aesthetic. 8000 credits with Kling gets you 800 generations if we just stick to the standard mode 5 second generations which is 4000 seconds or around 67 minutes of generation per month. That sounds good until you realize that you're going to end up generating 5 videos per one that will fit into your narrative well on the very low end and it could easily be 10 or even 20 if you're being discerning. So it's more like 4-12 minutes of footage if you're entirely focused on a single short film idea. For comparison, I probably generated 120-150 Gen-3 videos just yesterday and those are all 10 second generations. It seems like the 10 second Professional Mode might not be the way to go which gives you even fewer generations per credit but a lot of times even 10 seconds feels restrictive with videos frequently just getting to the good stuff around that point so being stuck with 5 second generations is going to give you even more of those situations where it had a great idea but just didn't have enough time to really develop it. The higher quality (at least with some prompts) might make it more attractive to the casual user and Gen-3 has no free option but it's hard to compare with how Gen-3 just lets you brute force your way to a usable generation without having to think about credits.
Oh yeah, the main reason i love Kling so much is because you actually get to try their free AI, which is fast like honestly dream-machine's main flaw is the generation time and their stupid queue system. They make me wait like i'm making a god damn 1 hour pixar movie. Its also accessible and credits refresh every day. It has proper image to video quality, might honestly even be the best one out there. Literally all they could have done right they have done right, and its at the worst its ever going to be from this point onward.
I don't understand what prompts I'm supposed to use for these allegedly amazing AI generators to actually produce something amazing. Every time I use one, it doesn't produce anything good at all. And this one has a horrendous mobile site. They obviously haven't made a proper site for using it on a phone. I put "a woman with brown braided pigtails using water-based magic to save a little girl from Roman soldiers" or something like that and it produced what looks like a woman talking to a little girl. That's it. And then I cut out the little girl part and it's just this chick splashing around in water.
you asked for 'juicy lemon', but there's no juice dripping from the lemon, on your hand, etc, and AI wasn't 'smart' enough, yet, to give you the pucker-face.
I’ve noticed from my experience with Kling Ai after running over 30 images it really does well in people eating and animals up close but for the weirder prompts naw it fails totally 👍. I think the model has only been trained for more of those people and animals. Not aliens,unknown creatures etc…
I wouldn't say the $5/month plan is unusable... it's for people who don't want to do too much, but want to have all the features available. I think there you mainly pay for the additional features because you only get 660 credits over the 66 per day you would get anyway without subscribing.
Thank you for this video! I would say that for English-speaking people, Luma and the new Runway Image-To-Video Gen 3 that just came out are both better than this. They understand English, this one understands Chinese. BTW Pixverse just recently unveiled their version of Luma. It is sometimes just like Luma and other times as bad as the old original flashy-cartoony AI videos. Very erratic. I will probably drop that since it's 30 dollars and get the 15 dollar version of this while it's on sale. I will try to use it for image to video that seems obvious, not the complicated or difficult prompts. Thanks again!
Just found out how costly and inefficient the credit system is. It uses up all credits instantly. Extremely overpriced. I thought this would be better than Runway. It seems these new AI have money as sole purpose of existence.
I’m finding the three biggies have different strengths and weaknesses! I’m doing a GOT’s type medieval project and Luma handles action best, although it does start off great then turn silly at times. Kling is good for big sweeping shots text to video but not so good with image to video. Runway has the best Image to video generations. Text to video misses the mark a lot though. I can’t say one is better than the other at this stage. Come on Sora, still waiting to be blown away!
Seriously tho irl when someone compliments a hair cut there usually just thinking it looked like shit before since they never complimented it before , compliments are blind insults
@@hipjoeroflmto4764You're really jumping to extremes with this one. You could think someone's hair normally looks like a 7 and make such comment on a day where it looks like a 10 to you. It doesn't mean it normally looked like a 1. You're assuming the worst unfortunately.
But is Kling cheap though? I think they are kinda greedy when it comes to generating in professional mode. 35 credits for a 5 sec video in pro mode is not cheap, especially when you have no idea about the outcome of the video you try to generate.
I wish it was subscription based an unlimited but maybe with usage limits of some kind in the sense that some way to protect over use. But I would love to hard use it in a day, but it would take a ton of gens to really get good at so its almost like throwing away some money if you want to expirament and get good with it. But perhaps for now it is fine, and surely in the future these things will be, possibly completely free. Well, likely completely free. Like everything else on earth.
AI stocks are set to dominate in 2024. I prefer NVIDIA because it is well-positioned to sustain long-term growth potential and offers a platform for other AI companies. I know someone who has seen returns of over 200% from NVIDIA. I'm also open to any other recommendations you may have.
I believe the next major breakthrough will be in A.I. For sustained growth similar to META, it's crucial to avoid making impulsive decisions based on short-term market fluctuations. Instead, prioritize patience and maintain a long-term perspective. Consider seeking financial advisory services to make informed buying and selling decisions.
@@Jonathancolter7100 In a similar situation, I sought advice from a financial advisor. By restructuring my portfolio and diversifying with good ETFs, the S&P 500, and growth stocks, I was able to grow my portfolio from $200,000 to over $800,000 in just a few years.
@@BrandonRichards21 Your advisor must be really good, I hope it's okay to inquire if you're still collaborating with the same advisor and how I can get in touch with them?
@@AlfredStephen127 I work with Claire Robert’s Durand as my fiduciary advisor. Simply look up the name. You would discover the information you needed to schedule an appointment.
It's impressive that they can make videos and stuff, but honestly, I don't think any of them are worth a paid plan at the moment, from what I have seen. The amount of "failed" generations will make it very expensive to make a video and even then, I don't think the quality is high enough for them to be useful for anything except a quick laugh. In my opinion, this technology probably needs at least another year or two before it is even remotely useful and not just a gimmick.
All this video did for me is advertise that better use runway. Like the whole video. you said: "kling is good" but you weren't impressed with it. And then you verified gen3 and you got better results and even said 'yeah it's better than kling'. But you kept pushing the idea that it's the "best generator". OVERALL I'd say both are bad. this feels like early versions of stable diffusion. Def not worth paying right now unless you want 'stock footage' but it's cheaper to just subscribe to a website that gives you plenty of stock images/videos. Def something for people who just want to have 'fun' and see at what stage we're currently in.
6:40 You 1 percenter can afford 3 prompts. 7:10 At your current price of 62 cents per 100, with 10 seconds costing 70 seconds, it would cost $312.48 to make a 2 hour movie, if every shot was perfect. 7:37 That's two videos in a row you used the word Ethereal. For those that went to public school like me - extremely delicate and light in a way that seems too perfect for this world.
is "officially the best" another way of saying "it sucks"? Because it definitely sucks. All 3 (luma, runway, kling) are overpriced garbage. They're fine for a few memes here and there but that's pretty much it. I know they'll get better with time, bur right now they're not worth wasting money on.
Bro these BOTS.... Will you spend any money on video generators just yet? or do they need higher quality?
Not yet, Maybe a year later
Pro or standard but unless I’m working on the major motion blockbuster $46 a month is ridiculous
None of them actually work yet reliably. I feel like Audio gen is miles ahead of video as far as consistency goes. Music hallucinations are there, but you get multiple takes to choose from and it rarely affects both. Videos come out with wild mutations, jump cuts to unrelated scenes, or they don’t animate at all and they burn credits. Even Kling has a large number of generations where I could only use the first 2 seconds out of 5. Luma is like 1-2 out of every 10 that are usable. Maybe Runway is better idk yet.
I prefer runway, I paid for both, I wish Kling had unlimited generations for premium then I'd pay for it. Btw runway released img 2 video
Lmao imagine spending money
11:18 rare footage of aliens mating in the forest
lol ikr
😂
Technically speaking that's not a zoom out. If that was done with a physical camera it would be a dolly out. The difference is that in an actual zoom out the relative positions of objects doesn't change.
They also got Pan and Tilt backwards (Pan = side to side , Tilt = up and down). "Horizontal" and "Vertical" have proper names as well (dolly and boom/crane/pedestal, respectively) but I get why they went with simplified terms for those.
Jeez Matt, can't you generate something photorealistic with real people doing normal things?
5:42 It sucks that camera movement is not available with ‘image to video’.
@@chariots8x230 agree
Very true
Matt GPT4o Voice Mode finally released! I hope you have access to it soon! Because i want to see it in action
trying to get access!
He won’t make a video on it precisely because you told him to, get shrecked, go back to 2012 L+Ratio
@@zrakonthekrakon494 do u suffer from any mental illness? Go get checked
You missed "last frame" function, which is one of the best uses for KLING.
After some testing with both paid Runway Gen-3 and Kling AI, I can confirm Kling AI is THE best video generator. It took an image of a voluptuous girl I created with SD about a year ago and it brought her to life for 5 seconds. It was amazing. I'm definitely going to be getting the subscription so I can continue the video up to 3 minutes!
Much like most text to video generators the magic is mostly in the image to video as an image has way more data for it to analyse and work from
Dont forget this is a 1 week half price discount, you have to count with the crossed out prices. And the Runway $95 tier has unlimited slow generations when all credits are utilized.
And slow mode isn't even that slow and still allows you to queue multiple videos at once
@@WehrtEuchEndlich really good point. That adds a lot of complexity to the comparison actually. Tbh I didn’t even know for kling that it was a week 1 discount. I thought it was a marketing tactic like temu 😭😭
Looks like they extended the 50% discount
@@roofoofighter Maybe Matt was right, it's just a tactic to create FOMO but it's never intended to be the supposed normal price. Still don't think it can compared to Gen-3 Unlimited if you're doing a ton of generating.
Professional mode has kept better face consistency for my use cases.
Its lucky when theres free ai text to video thatgives you daily credits for free
Gen3 just got the ability to upload a prompt pic
And while giving you 30 times fewer generations for a higher price
Difference between standard mode and professional mode?
Glad to see there's growing competition in this field!
Thanks for doing a price comparison with Runway, that was very useful given how much I got stung for the generation cost on that one
Anyone have a tutorial on how to use?
You missed the image to video to image end point . Good video I really wanted to see that feature if it was anything like Luna image to video to end point
The cat gens and the octopus were my favorites, it'd be so cool if they could add atmospheric sound effects to fit the video
Kling is good if you want realism. If you want to do cartoon or drawing styles, Dream Machine by Luna AI is better. I hope RunwayML gives image to video with Gen 3 soon.
Matt eating a lemon is pure nightmare fuel!
I can only imagine what AI Video generators will be like in the future...
Unfortunately, I bought a $35 Gen3 subscription, and I am disappointed. I’m going to switch to Kling-it’s a no-brainer.
I feel like Kling is specifaly especialy trained on the stuff they showed in their promos to cheat a little bit
Everytime I upload a photo with some sort of piano (even a closed e piano),
it tends to just add more of the acoustic stand up pianos shown in the promos
as soon as the camera moves a bit to the left or right a bit.
Also Kling is way better at generating pandas, than monkeys for example
That explain the their excellent eating video given all the mukbang datasets. 😂
They need to add end frame feature just like for what luma dream machine AI has. Kling ai and runway 3 definitely need to add this cool feature in their next update hopefully.
Thank you for the tutorial. Were you able to extend any of your videos to 3 minutes? What about the end frame functionality?
Not sure from their terms if we can use this for commercial/client projects
doesn't mention anything about commercial use, do you have any insight on this?
So I made a couple image to videos just now... prompt following was terrible but what was generated looked ok quality wise...
I wish there was camera control on image to video, they could easily animate things if an image is given, like Leonardo AI. But despite more information content from the image, why isn't camera available? Also there is an extremely great need for higher quality, and resolution selection on image to video. But overall I would prefer Kling over Gen 3 because Gen 3 is supremely expensive and inefficient.
How come Genmo AI have default almost 2K high quality videos but other AIs do not? Even if Kling and others do add high quality options, they would make them extremely costly now. Wish they had default 2K quality like Genmo AI does.
Maybe running the prompts through a LLM for a translation would help out?
Well an llm might not be needed for that just the best translator, I've heard great things about DeepL
I hope we will have some sort of physics sim model inside these generators for real use soonish.
Matt is there a way for us to download the source to any of these AI to video, also which AI to Video generators allow you to keep generating one after another of the same project? Someone needs to start an open source free AI to video Editing Software for desktop like davinci resolve.
Matt, you been watching Cops again??
I still think Gen-3 is the best option for serious creators if only for its unlimited plan. Making anything presentable with these tools takes a lot of generating and tinkering with prompts so you are going to burn through a ton of generations trying to put together something quality with a cohesive aesthetic.
8000 credits with Kling gets you 800 generations if we just stick to the standard mode 5 second generations which is 4000 seconds or around 67 minutes of generation per month. That sounds good until you realize that you're going to end up generating 5 videos per one that will fit into your narrative well on the very low end and it could easily be 10 or even 20 if you're being discerning. So it's more like 4-12 minutes of footage if you're entirely focused on a single short film idea.
For comparison, I probably generated 120-150 Gen-3 videos just yesterday and those are all 10 second generations. It seems like the 10 second Professional Mode might not be the way to go which gives you even fewer generations per credit but a lot of times even 10 seconds feels restrictive with videos frequently just getting to the good stuff around that point so being stuck with 5 second generations is going to give you even more of those situations where it had a great idea but just didn't have enough time to really develop it.
The higher quality (at least with some prompts) might make it more attractive to the casual user and Gen-3 has no free option but it's hard to compare with how Gen-3 just lets you brute force your way to a usable generation without having to think about credits.
Thanks Matt, will check this out. Hair looks fab 😎
good stuff. mostly been using luma. will try out kling professional mode too
obviously these features or more will appear in Runway 3...eventually..
How do you figure this is the best video generator? Or was that just a click-bait title.
in image to video luma still the best. Zero morphing in my expierence.
Next week? how about tomorrow
Oh yeah, the main reason i love Kling so much is because you actually get to try their free AI, which is fast like honestly dream-machine's main flaw is the generation time and their stupid queue system. They make me wait like i'm making a god damn 1 hour pixar movie. Its also accessible and credits refresh every day. It has proper image to video quality, might honestly even be the best one out there. Literally all they could have done right they have done right, and its at the worst its ever going to be from this point onward.
What about the 3 minutes video feature?
What happens if credits doesn’t renew while expires within 24 hours. I don’t care if i buy plans after all
I forgot the answer is. Actually it is. When next day it will receive daily credits
5$ is nice for a lot of hobbyist
Bro you look tired af, go to sleep! 😂😂
I already sign up for the 1 month high price if i think i will get the 6 months plan before the deal ends😅
18:06 wait did he just say edging
I wish you tested out the high quality 10 second generations more on different images and then the extend feature.
@@thegreatdelusion you’re right. Maybe on a livestream?
I don't understand what prompts I'm supposed to use for these allegedly amazing AI generators to actually produce something amazing. Every time I use one, it doesn't produce anything good at all. And this one has a horrendous mobile site. They obviously haven't made a proper site for using it on a phone.
I put "a woman with brown braided pigtails using water-based magic to save a little girl from Roman soldiers" or something like that and it produced what looks like a woman talking to a little girl. That's it. And then I cut out the little girl part and it's just this chick splashing around in water.
you asked for 'juicy lemon', but there's no juice dripping from the lemon, on your hand, etc, and AI wasn't 'smart' enough, yet, to give you the pucker-face.
I’ve noticed from my experience with Kling Ai after running over 30 images it really does well in people eating and animals up close but for the weirder prompts naw it fails totally 👍. I think the model has only been trained for more of those people and animals. Not aliens,unknown creatures etc…
16:55 That’s hella cursed
Broken objects were always difficult to do for AI. Best image generators can't do this properly as well.
I wouldn't say the $5/month plan is unusable... it's for people who don't want to do too much, but want to have all the features available. I think there you mainly pay for the additional features because you only get 660 credits over the 66 per day you would get anyway without subscribing.
So basically the extend video to 3 minutes doesn't work right?
I finally subscribed today 😅 ! Being a youtuber is hard these days
The amount of rule 34 and new hentai content this tech will generate in the coming decades will be absolutely diabolical.
Thank you for this video! I would say that for English-speaking people, Luma and the new Runway Image-To-Video Gen 3 that just came out are both better than this. They understand English, this one understands Chinese. BTW Pixverse just recently unveiled their version of Luma. It is sometimes just like Luma and other times as bad as the old original flashy-cartoony AI videos. Very erratic. I will probably drop that since it's 30 dollars and get the 15 dollar version of this while it's on sale. I will try to use it for image to video that seems obvious, not the complicated or difficult prompts. Thanks again!
As a video guy, it qute frustrates me me that none of the creators of these ai generators know what zoom is.
Just found out how costly and inefficient the credit system is. It uses up all credits instantly. Extremely overpriced. I thought this would be better than Runway. It seems these new AI have money as sole purpose of existence.
NOT IMPRESSED - At this point it falls short on everything. W;ould never put my money in any of these AI platforms. Too early!!!
the exact same discount popup as on Temu and Alibaba..those chinese are ramping up for war
seems still soo far from Sora demos (but of course how real/representative are those...)
Does anyone know how to remove the watermark from Kling? The plans say you can remove the watermark but I don’t see an option.
Just choose “Download without watermark”. This option appears, when you move your cursor to download button before clicking on it.
@@aee0320 I’m on my mobile so that option doesn’t seem to exist.
I’m finding the three biggies have different strengths and weaknesses! I’m doing a GOT’s type medieval project and Luma handles action best, although it does start off great then turn silly at times. Kling is good for big sweeping shots text to video but not so good with image to video. Runway has the best Image to video generations. Text to video misses the mark a lot though. I can’t say one is better than the other at this stage. Come on Sora, still waiting to be blown away!
The “tilt” setting is a “pan”. Lost in translation?
odd idea i had is that maybe AI just feels nervous about making things blow up :p
Your hair is looking good! LoL
In other words? "Your hair looked like shit before!"
Seriously tho irl when someone compliments a hair cut there usually just thinking it looked like shit before since they never complimented it before , compliments are blind insults
@@hipjoeroflmto4764You're really jumping to extremes with this one. You could think someone's hair normally looks like a 7 and make such comment on a day where it looks like a 10 to you. It doesn't mean it normally looked like a 1. You're assuming the worst unfortunately.
@@BarryMcBangerz then why wouldn't they compliment at the 7 then?
@@BarryMcBangerz why do people wait for change before telling someone how they feel
I don't hear much about meta ai image animator much. It's not bad for free.
It's still bad though
But is Kling cheap though? I think they are kinda greedy when it comes to generating in professional mode. 35 credits for a 5 sec video in pro mode is not cheap, especially when you have no idea about the outcome of the video you try to generate.
Why does everything have to be cheap?
Bro.... you have a million Gen3 credits.... you can start a Gen Credit-Credit Agency! let me borrow 500,000 im good for it i swear
Great video.
Kling ai, I tried but it needs a lot of improvement
I wish it was subscription based an unlimited but maybe with usage limits of some kind in the sense that some way to protect over use. But I would love to hard use it in a day, but it would take a ton of gens to really get good at so its almost like throwing away some money if you want to expirament and get good with it. But perhaps for now it is fine, and surely in the future these things will be, possibly completely free. Well, likely completely free. Like everything else on earth.
And how much did they pay you?
i'll just wait. thanks for saving my money for now 🙃
why is bro oiled up in this one
Lol? Why does adding 10 seconds add on 60 credits to it , please somone tell me why that makes sense and its not a scam
AI stocks are set to dominate in 2024. I prefer NVIDIA because it is well-positioned to sustain long-term growth potential and offers a platform for other AI companies. I know someone who has seen returns of over 200% from NVIDIA. I'm also open to any other recommendations you may have.
I believe the next major breakthrough will be in A.I. For sustained growth similar to META, it's crucial to avoid making impulsive decisions based on short-term market fluctuations. Instead, prioritize patience and maintain a long-term perspective. Consider seeking financial advisory services to make informed buying and selling decisions.
@@Jonathancolter7100 In a similar situation, I sought advice from a financial advisor. By restructuring my portfolio and diversifying with good ETFs, the S&P 500, and growth stocks, I was able to grow my portfolio from $200,000 to over $800,000 in just a few years.
@@BrandonRichards21 Your advisor must be really good, I hope it's okay to inquire if you're still collaborating with the same advisor and how I can get in touch with them?
@@AlfredStephen127 I work with Claire Robert’s Durand as my fiduciary advisor. Simply look up the name. You would discover the information you needed to schedule an appointment.
It's impressive that they can make videos and stuff, but honestly, I don't think any of them are worth a paid plan at the moment, from what I have seen. The amount of "failed" generations will make it very expensive to make a video and even then, I don't think the quality is high enough for them to be useful for anything except a quick laugh. In my opinion, this technology probably needs at least another year or two before it is even remotely useful and not just a gimmick.
Vidu ainow available to all
wow. you just got out of the sauna. that is not fresh.
❤
All this video did for me is advertise that better use runway. Like the whole video. you said: "kling is good" but you weren't impressed with it. And then you verified gen3 and you got better results and even said 'yeah it's better than kling'. But you kept pushing the idea that it's the "best generator".
OVERALL I'd say both are bad. this feels like early versions of stable diffusion. Def not worth paying right now unless you want 'stock footage' but it's cheaper to just subscribe to a website that gives you plenty of stock images/videos. Def something for people who just want to have 'fun' and see at what stage we're currently in.
It works better if you use ChatGpt or Claud to translate your prompts into Beijing style Chinese ... Don't use Google translate, it sucks.
testet alot LumaAi is better quality.
Bro these videos are so low effort. Would it kill you to record and actually edit some side by sides shots?
6:40 You 1 percenter can afford 3 prompts. 7:10 At your current price of 62 cents per 100, with 10 seconds costing 70 seconds, it would cost $312.48 to make a 2 hour movie, if every shot was perfect. 7:37 That's two videos in a row you used the word Ethereal. For those that went to public school like me - extremely delicate and light in a way that seems too perfect for this world.
35 credits for "Professional Mode" for a Chinese image to video generator 🤣😂😅
I know this is unrelated but give the Quran a read.
Is that a new model
Why is bro so sweating
is "officially the best" another way of saying "it sucks"? Because it definitely sucks. All 3 (luma, runway, kling) are overpriced garbage. They're fine for a few memes here and there but that's pretty much it. I know they'll get better with time, bur right now they're not worth wasting money on.
А
B
Try this prompt "Winnie the Pooh waving the flag of Taiwan" see if the CCP is watching or not.
🫡