วีดีโอ

@gamingguruoz Veo 2 certainly is a model we can't wait to examine thoroughly :)

Thanks for your comment, but this video review is about Hunyuan, not Hailuo.

Hailuo's I2V model needs refinement, for sure, but we had a much better experience with it than you. Hard to guess, but try simple action in your shots, as it gets sometimes when the movements within the frame are a tad too complex. We haven't used their customer service yet, so we cannot confirm or deny if it is good or bad.

Hi! Not yet, we're afraid. We need to wait - we're sure they're working on it as we speak.

so what are you settling on? I have been researching for 2 weeks before i purchase one. Runway, minimax, kling?

$1400??? Are you sure you're talking about Kling AI? The last time we checked (about....5 minutes ago), the Pro Plan was still priced $32 or so /month.

unlikely

Lio==[]::::::::::::::::>o==[]::::::::::::::::

Glad you liked it!

Thanx! We plan to make a video about lipsync, sound, and a film score. The same goes for camera movements, so stick around :)

Que bom que você gostou! Estamos focados em ensinar cinematografia real, apenas usando ferramentas de IA em vez de câmeras tradicionais. Aproveite o resto dos vídeos!

Hi! Why would you say that? Have you watched our videos reviewing Hailuo t2v and i2v features? To say that we "hate" MiniMax couldn't be further from the truth.

So nice of you! We'll do :)

Thank you, highly appreciated!

Thanks a million!

Thanks! We're trying to save your precious time, so you gotta rewind :D

Hi! Yes, it looks promising. Don't worry about problems accessing Sora - even in the US, there are problems with opening an account.

Hi! This video was NOT sponsored by anyone - we'd inform the public if it was. We chose this topic because we thought it might be interesting for our subscribers.

we know which one he hates here so t=you know hes bias t bottom left hand one

We're considering a review of Vidu and a bunch of other generators, so stick with us :)

@@AIFilmmakingAcademy Great! Thanks Sir.. ☕

The leaks are promising, but let's wait till we can seriously test it. That's a risky strategy to create so much hype - now, anything short of breathtaking will likely lead to disappointment.

We do, and we will be using it :) And it will get better and better.

I do

AI is the future, and the future is here.

Marques just released his review of SORA and yes it does have image to video

Sora will have all: i2v, t2v, and v2v. And yes, the list of off-limits themes is long, so we may end up producing sweet videos of lovely cats only ;)

@@AIFilmmakingAcademy Lol, $200 p/m to make stock footage, oh lordy!

Yep, the pricing seems to be quite steep for non-pro users. With the basic plan you cannot e.g. upload images of people to make i2v - you'd need to fork out $200 for that.

Actually, Freepik is quite an interesting option nowadays.

@@AIFilmmakingAcademy I was surprised to see those options there

Welcome to the channel! Thank you for your comment, we appreciate it!

Thanks for the tip! Makes sense, particularly for those who want to dedicate themselves to open-source models.

100%. I just ordered one such out of my tech collection weakness and then thinking to myself; But I don't really have any tasks that will drive this to its worth.. and then this video popped up on my timeline. I believe the algorithm followed my purchase and linked to your comment. :D

I have been thinking about this too, but not sure if you can use your gpu in your computer plus the gpu on the external and combine the vram, I have a rtx 4080 super with 16 g of vram, but they talk about this program need up to 80 g of vram.

@leanfwc it depends if the software supports multiple gpus but I hope that it does. On eBay they have NVIDIA Tesla K80 24GB GDDR5 GPU Accelerator Graphics Card which would bring my system up to the 40 gig minimum capacity.

@@HistoryViper You can go for the NVIDIA A100 80GB (HBM2e) hehehe.. Then you have even over 2 TB per second of memory bandwidth and can run your own server with it as well if you go for the SMX version

Thank you!

@@AIFilmmakingAcademy Totally SURE! really useful video tutorial! Let's do best.. have a good day with coffee everyday. ☕

You too!!!

Glad you liked it!

Hi! Thanks for your kind words - much appreciated. We have a video on SFX in AI filmmaking on our "To Do" list :)

Hi! Thanks for your comment. That's a valid question. Certainly, starting with basic monthly subscriptions seems like a good option to give yourself the time to try them and make up your mind. The problem is that - at least some of the platforms - offer much better quality and more functions in the "professional" modes. We can recommend watching reviews on YT to find out if these professional modes are worth their price. Just FYI - whatever we test, we always have the pro subscription. Where it is appropriate, we explicitly say in the video that certain functions or features are available only in the "pro" mode. Enjoy the rest of the videos! A new one should be published tomorrow :)

Hi! We're not sure we understand your point...

Thanks for watching and welcome! Enjoy!

Hi! Yes, you can remove it with any paid plan, I guess. We use the Unlimited Plan, so we could've removed it, but we left it so the viewers know which AI video generator we used to create the examples.

Do you mean "how long does it take to make a video"?

We'd disagree here. You can already make approx. 20-second long long-shots. As you can see, the acceleration in AI generators development is mind-blowing. And....AI does not have to predict anything, it just needs to follow your prompt so it's tough to compare it to a weather forecast.

Thank you!

Absolutely! True, especially when what you want from AI is more sophisticated and non-standard. Thanks for your comment!

Welcome to the channel! Thank you, much appreciated! Enjoy the rest of our tutorials and reviews.

@@AIFilmmakingAcademy Sir, could you reviews my music-video AI? i give you permission. i'm your permanent subscriber. honest, you can reaction or analyze good part and bad part of my video or another subscriber video on your channel. if not it's ok. because, i see every video you share too much "VITAMIN" that can inspire all viewers. not ttoo much talking and to the point. it's so high class. thanks sir!

We'd be happy to, but please allow us some time as we're honestly totally busy nowadays :)

@@AIFilmmakingAcademy Ok . . Sir.. i'm waiting your surge and your every video that inspiring me for all other creators.. for my progress i try to create mike tyson vs jake paul in ai with your dutch skill.. thanks Sir.. have a good day..

@@ErwinWijayantoEntertainment All the best, enjoy AI filmmaking, it's so much fun!

Thanks for your comment. Yes, 720p will be very much welcome :) And the technical requirements for training Genmo are a stretch, indeed.

Hi! Welcome to the channel. You can check the options Genmo offers here: www.genmo.ai/blog

Hi! Is this a question or a statement?

@AIFilmmakingAcademy a statement. Prompt to vid is really good for one offs and a bit of fun. But creating a cohesive video wirh cut after cut then it falls short. 1. If u generate a image and dont like it then you can redo. Most image generators give you 1000s a month for 10 dollars. 2. Videos take ages to make and if you typing in words you dont know what look or vibe your going to get, could be a waste of time and resources. Hope they add it soon.

OK, now we understand :) Certainly, the image-to-video or video-to-video models make it easier to achieve consistency, especially when you need main characters to remain consistent. We can think of a few applications of the text-to-image models, though. Of course, we are awaiting the image-to-video functionality, too. BTW, welcome to the channel!

😊 Was just having a rant in the comment section... Yes hopefully we can have 1. Image to vid. 2. Top up of credits. No having a top up is also really bad because a 5 min vid could easily eat up 100+ generations in a day. 😢.

@@armondtanz We do sometimes wonder why more AI video generators do not offer unlimited plans, like Runway does. It might be based on a fear that some bots can eat up the system capacity by generating tons of useless video and clogging the generator.

Thanks so much, as always, Carlos! Glad you enjoyed it :)

The video I generated looks very good, except that the subject's head (a woman) was briefly partially transparent where you could see through the top portion of her head. 😅 Very odd, as I've never seen that happen with any other A.I. video generator result, but other than that the result looked great.

This is exactly our impression - you can generate a great shot with stunning camera movement and decent subject's motion, but it's ruined by some morphing on the part of the hero's body, or an infamous 6-finger palm, or whatever. We hope the engine is now being fed tons of generations, so the learning curve should be quite steep. It should help fix these flaws. We're definitely gonna keep an eye on Genmo :)

Thanks for your comment. Can you elaborate on this?

Hi! Thanks for your comment. Each of the shots in this video was done by us. We typically make one or two attempts only for each prompt. This video was not sponsored by Genmo, and we did not use any of their promotional materials. What you see in the video is what you can achieve by yourself using Genmo Mochi 1.

@@AIFilmmakingAcademy ok

@@AIFilmmakingAcademy Although I don't know what the situation is now because I tried this AI 3 months ago last time.

So, we believe you tried the old model, as Mochi 1 (the one we reviewed in the vide) was released 5 or 6 DAYS ago. It's a totally new one.

@@AIFilmmakingAcademy Yes indeed it was an old model then. Now I have tried the new model and it is better.

With all top AI video generators, it is still hit or miss sometimes. We use all top 4 (Kling, Runway, Luma, and MiniMax) for some projects - certain shots just come out better/quicker with one or the other, and there is no clear-cut specialty, like A is great for people running while B is great for aerial shots. You're 100% right about Runway's unlimited option - it is something we'd expect from the other 3 companies to offer as an option.

HI! Welcome to the channel! Thank you for your kind words. Please enjoy the rest of the videos. New content is coming soon :)

Hi! Thanks for your comment. Consistency is the main issue we're all facing. We can share two tips: 1) use the image-to-video function (and create consistent images in your preferred text-to-image generator). 2) Be precise in your prompts and describe not only the action but also the set design elements to maintain consistency. We plan to do a separate video on consistency so stay tuned!

@@AIFilmmakingAcademy thanks. Please pick a story of of at least 6 scenes, having different and multiple characters in scene also the place or house or cafe with shots of different camera angles. I’m doin so using last frame or mid frame to next scene for consistency.

@@AIFilmmakingAcademy also please if you can upload a video combining shots of different camera angles along with the prompts as you showed a couple prompts and shots in this video which were completely new for me and improved my videos generated using hailuoai, will be really very helpful for us to learn. Many thanks,

Thank you! Cheers!

Glad you enjoyed it!