AI Text-To-Video Is Here! Install It On Your PC NOW!
Рет қаралды 38,728
Жыл бұрынExperience the magic of AI-powered Text-to-Video in action! Modelscope ai model just came out where you can enter your text prompt and witness how it generates a video from text in a minute! In this video, I will show you how to use the huggingface demo, clone it in your private space, then how to use it on your own computer using the brand new auto1111 extension and how to remove the shutterstock watermark from a video! So let's go!
Did you manage to make a video from text using the ai? Let me know in the comments!
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
SOCIAL MEDIA LINKS!
✨ Support my work on Patreon: / aitrepreneur
⚔️ Join the Discord server: bit.ly/aitdiscord
🧠 My Second Channel THE MAKER LAIR: bit.ly/themakerlair
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Free Demo: huggingface.co/spaces/damo-vi...
Models: huggingface.co/damo-vilab/mod...
Extension: github.com/deforum-art/sd-web...
Watermark remover: anieraser.media.io/app/
Original repo: modelscope.cn/models/damo/tex...
INSTALL Stable Diffusion Automatic1111: • FREE 2023 Stable Diffu...
All ControlNet Videos: • ControlNet
My previous ControlNet video: • TRANSFER STYLE FROM An...
3D POSE & HANDS INSIDE Stable Diffusion! Posex & Depth Map: • 3D POSE & HANDS INSIDE...
Multiple Characters With LATENT COUPLE: • MULTIPLE CHARACTERS In...
GET PERFECT HANDS With MULTI-CONTROLNET & 3D BLENDER: • GET PERFECT HANDS With...
NEXT-GEN MULTI-CONTROLNET INPAINTING: • NEXT-GEN MULTI-CONTROL...
CHARACTER TURNAROUND In Stable Diffusion: • CHARACTER TURNAROUND I...
EASY POSING FOR CONTROLNET : • EASY POSING FOR CONTRO...
3D Posing With ControlNet: • 3D POSING For PERFECT ...
My first ControlNet video: • NEXT-GEN NEW IMG2IMG I...
Special thanks to Royal Emperor:
- Totoro
Thank you so much for your support on Patreon! You are truly a glory to behold! Your generosity is immense, and it means the world to me. Thank you for helping me keep the lights on and the content flowing. Thank you very much!
#stablediffusion #AI #TextToVideo #VideoCreation
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
WATCH MY MOST POPULAR VIDEOS:
RECOMMENDED WATCHING - My "Stable Diffusion" Playlist:
►► bit.ly/stablediffusion
RECOMMENDED WATCHING - My "Tutorial" Playlist:
►► bit.ly/TuTPlaylist
Disclosure: Bear in mind that some of the links in this post are affiliate links and if you go through them to make a purchase I will earn a commission. Keep in mind that I link these companies and their products because of their quality and not because of the commission I receive from your purchases. The decision is yours, and whether or not you decide to buy something is completely up to you.
@Aitrepreneur Жыл бұрын
HELLO HUMANS! Thank you for watching & do NOT forget to LIKE and SUBSCRIBE For More Ai Updates. Thx
@Anonymous-gu2pk Жыл бұрын
Other viewers may not realize this, but this creates a good foundation for img2img batch edit (which also gets rid of the watermark). You can use ControlNet for better consistency and I think you could reduce the flickering with DaVinci Resolve.
@michaelwoods7770 Жыл бұрын
is there a way for me to run it with my own hardware have a 3090 which should be enough to do so. cpu:5800x3d 32gb ram.
@MatichekYoutube Жыл бұрын
@@michaelwoods7770 yes, that would be more than enough
@jameshughes3014 Жыл бұрын
it boggles my mind that this runs on anything less than an A100. Who ever put this together did a great job.
@lifemarketing9876 Жыл бұрын
10 years from now we will be able to create photorealistic full length movies wth text. This is only the beginning.
@Myplaylist892 5 ай бұрын
@@jeanmichel8919 YES
@SoCalGuitarist Жыл бұрын
Literally installing the new A1111 Extension for this right now! Awesome. Also "What a time to be alive!" - Love that 2 min papers callback 😉😉
@Aitrepreneur Жыл бұрын
He's the ai youtuber pioneer after all :D
@ryry9780 Жыл бұрын
Vram required for this?
@vi6ddarkking Жыл бұрын
Hopefully some of this tech can also be adapted into text to image in the future in stable diffusion in order to FINALLY get consistent characters and images.
@iamisobe Жыл бұрын
ya character consistency in stable or midjourney is so hard it sucks, im trying to do a comic book and while the characters are similar I still feel theyre different enough itd be off
@Anonymous-gu2pk Жыл бұрын
Just a minor suggestion: you can also download the files by just clicking the little downwards arrow for each file on the download page.
@TTTrouble Жыл бұрын
lol honestly, I knew there should be an easier way but can't tell you how mildly irritated I've been the last few months needing to go into click more times than I felt was necessary to download some of these models. Thanks for the tip, cheers friend!
@NegativeFiveee Ай бұрын
You teach me everything I know. I love you. Dont let anyone take that away form us.
@WellAnimated Жыл бұрын
"What a time to be alive!" with the accent xD
@Aitrepreneur Жыл бұрын
😁
@trippymccube8735 Жыл бұрын
hahaha hold on to your papers!
@Orangesnake221 Жыл бұрын
Exiting times! This almost reminds me of the first day stable diffusion! Give it two or three months and this will be perfect!
@gastonslashx Жыл бұрын
What a time to be alive! ha, a real two minute papers moment right there, love it.
@RetzyWilliams Жыл бұрын
Have it working thx so much! For Win keep eye on Command prompt in background, when you click "Generate" it looks like nothing is happening, but in Command prompt you see the processing.
@Aitrepreneur Жыл бұрын
Good tip!
@jetson35 Жыл бұрын
i made a comment on the reddit post to "wake me up when its an extention for stable diffusion" i didnt actually think they were gonna make one over night lmfao
@Ez-se2dl Жыл бұрын
Nice video! Waiting for the next one about Gen-2 :)
@ryry9780 Жыл бұрын
Time for Corridor Digital to make episode 2 with this now 😂
@HolidayAtHome Жыл бұрын
You indeed are my AI Overlord !
@ayoubachak2154 Жыл бұрын
Great content G
@traceurAlex Жыл бұрын
What a time to be alive!! Reference!
@johncloud998 Жыл бұрын
can't wait !!!
@TrueInvisible Жыл бұрын
soon enough.. we'll be making A.I full length films that will drive the creatively bankrupt cinema out of business!
@CHARIOTangler Жыл бұрын
People will probably just make idiotic Transformers and Fast and Furious reboots. I don't expect any modern-day Holy Mountains🤣🤣🤣.
@TrueInvisible Жыл бұрын
@@CHARIOTangler .... i DO realize the tidal wave of deranged western degenerate production, but i also have hopes for actual creativity to take advantage of this.. we already established that the west if officially creatively disabled, bankrupt, and just a milking cow by this point. also it's violating and defiling older franchises, that's why i pray that this technology will be used by the creative indies and the actually gifted artists.
@Anonymous-gu2pk Жыл бұрын
A large corporation with massive financial, technological and human resources will always be able to use the same tools to create something greater than a single individual can. The only question is which company is willing to do it and which one would rather crash and burn.
@TrueInvisible Жыл бұрын
@@Anonymous-gu2pk from what i can see nowadays ... "large corporations with limitless financial and technological and disposable trash [humans] resources" choose to "make" TRASH for products and milk them for money using rage and hatred reactions. instead of actually doing anything creative at all with them. they maybe strong, resourceful and advanced, but "innovation wise" they don't exist.
@TheMemeSong Жыл бұрын
Why not? We can make birdermic sequel with that!!!
@darkesco Жыл бұрын
Best AI channel on youtube.
@InFlamesOfSorrow Жыл бұрын
oooh i am just waiting for a combination between modelscope text to video and controlnet with open pose + hands(canny +depth) via Blender. Should be able to create some awsome animations that are way less funky
@tdfilmstudio Жыл бұрын
Excited!!! Please Guide us
@flonixcorn Жыл бұрын
Nice!
@godned74 Жыл бұрын
Thanks for putting the extra links because it wasn't showing up on my webui.
@iamYork_ Жыл бұрын
Just be careful when explaining how to remove watermarks... In theory that is illegal activity and I love your channel i dont want you getting anymore strikes...
@electrolab2624 Жыл бұрын
Thx so much! - Mine did not give the error and created a video you can display on the Auto1111 extension. A good sign!
@MathewRenfro Жыл бұрын
a "Remove-shutterstock-logo" template mask would come in handy, to inpaint out the shuterstock logo.
@alus0l Жыл бұрын
are you the one who does 2Minute Papers? What a time to be alive!
@robynhighart2026 Жыл бұрын
I knew it. I predicted this over a couple years ago. I'll be able to plug in a book, adjust some parameters and have a full movie made of it.
@imanonymus5745 Жыл бұрын
Just few hours ago runway launched their ai to video
@sambas9257 Жыл бұрын
At this moment it's really perfect for viral videos and animated memes/emoticons. Geek tik-tokers will be crazy with this thing ...
@RAKIBKHAN-gm9dx Жыл бұрын
Runwayml also launched text to video try it also
@MarcoKatITA Жыл бұрын
can i use my personal ckpt file on hugging face online ? my pc is not powerfull enough and runpod is too slow for import all the 10 gb files inside jupyter lab
@FusionDeveloper Жыл бұрын
my extensions URL is different and it doesn't list that extension. I changed mine to yours and it automatically changed to: {} and gave me zero results and an error. Maybe i need to update some other automatic11111 files, but I went ahead and just chose to install from URL and put in the github URL and it installed and worked.
@yumri4 Жыл бұрын
How different is it from the demo on huggyface ? I currently don't have the hardware to use it but huggyface has a demo that i can use so i want to know if i can expect it to output the same or similar videos from text or is it very different when ran on the full AI instead of the demo version?
@greendsnow Жыл бұрын
you little hacker you! :D
@evilpenguin99999 Жыл бұрын
CUDA out of memory. Tried to allocate 114.00 MiB (GPU 0; 8.00 GiB total capacity; 7.14 GiB already allocated; 0 bytes free; 7.29 GiB reserved in total by PyTorch) fml cant use it.
@FableFrenzy Жыл бұрын
Ai please generate a video of Shark Nado but make it a 10/10
@Anonymous-gu2pk Жыл бұрын
"8gbs of VRAM on top of SD should be enough to launch (when the VAE unloading will be fixed, *before that orient around ~12 gbs)."*
@leemark7739 Жыл бұрын
what does it mean
@Anonymous-gu2pk Жыл бұрын
@@leemark7739 It could be interpreted a couple of different ways, but it sounds like they were initially suggesting that you'd need at least 12GB VRAM to run the extension. That's why people with 8GB couldn't do it. There have been some updates now and I've been able to run it with 10GB. I don't know what's the latest news for people with 8GB VRAM.
@leemark7739 Жыл бұрын
@@Anonymous-gu2pk do we have any other ways to run it with 8gb
@Anonymous-gu2pk Жыл бұрын
@@leemark7739 I don't know... the updated version has an option to run with CPU, but that's probably quite slow. There's also the slow option online that's linked as free demo in the video description. And the video mentions paid options online.
@gibsonfan-sm6nw Жыл бұрын
@@leemark7739 I am able to run it on 8gb although only at a resolution of 256x256. I am going to play around with img2img/upscaling to see if I can get better results as well.
@grape7569 Жыл бұрын
can we generate the animation without background?
@ractorstudios Жыл бұрын
now we just need midjourney style videos
@Anonymous-gu2pk Жыл бұрын
Run the generated images through img2img with OpenJourney model and ControlNet extension.
@gibbo271 Жыл бұрын
can you make a tutorial to use this paired with a discord bot?
@trendqiang3921 Жыл бұрын
👍
@MannyLectro Жыл бұрын
I'm getting a cuda out of memory error no matter what I do, even if I check the low vram box. I have a 2080 super and usually don't have any issue with vram. What am I missing?
@evilpenguin99999 Жыл бұрын
Same, 1080 with 8GB RAM in my case.
@leemark7739 Жыл бұрын
me too 3060ti with 8 ram
@flow-saf Жыл бұрын
Did I hear "What a time to be alive"? 😅
@AbsurdShark Жыл бұрын
Hopefully Shutterstock won't sue them.
@mr.justdoev8245 Жыл бұрын
shutterstock ? no nsfw ?? :/
@user-rl6te5my8d Жыл бұрын
i always gets GPU failed error after few attempts "try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF" can you please make video how to fix it
@leemark7739 Жыл бұрын
same error
@Grunacho Жыл бұрын
Thank you for the video! Do someone know, why I experience python crashes after a while.(Win11 x64) However, I can do some videos without crashes. It cannot 100% reproduce the error. My PC is powerful enough. I use Python 3.10.6. I tested also 3.10.9 without success. Thanks ;-)
@geyck Жыл бұрын
Yeah, I'll wait for this one. "Here's a hack to pirate the pirating software!"
Жыл бұрын
Károly is that you? 🧐😁
@Aitrepreneur Жыл бұрын
It's DR K !
@jetson35 Жыл бұрын
getting a Exception occured 'NoneType' object has no attribute 'cond_stage_model' error, anyone find a fix? edit: so it went away after i restarted the UI but now i get a constant low memory error even if i use cpu (i have 8gb of vram)
@TurbineFlyer Жыл бұрын
Wow. I have a 3070 and it keeps running out of memory.. RIP
@wazaN Жыл бұрын
I wouldn't mind if you brought more criticism because let's face it. It's not good enough ! No one is going to use it like we use stable diffusion right now. Of course it's a progress, but there is still a looot of work to do.
@Aitrepreneur Жыл бұрын
First step toward the goal is often gonna be bad ;)
@jimlthor Жыл бұрын
If you've seen how far the text to image stuff has come in a very short time, you'd understand why this is exciting.
@elviskabwe3639 10 ай бұрын
how can i run this on my pc and start creating my own
@ireincarnatedasacontentcreator Жыл бұрын
how about imagetovideo on SD?
@lowserver2 Жыл бұрын
any way to make a video from an image instead of text?
@jimlthor Жыл бұрын
Looks like there's another extension called SD something. I havent tried it yet, but that's what the description says.
@Nakamako1 Жыл бұрын
not good enough yet unless we can use adons
@kennylex Жыл бұрын
I am not so smart so I often become suspect, so why do it do "Git commit", is that not the command for Git to send something to a server?
@tetsuooshima832 Жыл бұрын
Well, it worked ONE TIME, then never again. Getting error "Exception occured 'NoneType' object has no attribute 'cond_stage_model', and I don't want to play the beta tester so I'll just delete it and wait for a final version. Let the true nerds fix it lol
@kimsteinhaug Жыл бұрын
I couldn't get the "hack" to work, did they already fix it ? Strange they were using cookies in the first place for this check.
@Aitrepreneur Жыл бұрын
I worked for me before but you can use another account and try again
@RicardoSilvaptl Жыл бұрын
the link for the model is in chinese, there is any EN version available?
@Aitrepreneur Жыл бұрын
I've updated the description ;)
@RicardoSilvaptl Жыл бұрын
@@Aitrepreneur thanks
@justinnine4940 Жыл бұрын
Can I animate my own dog?
@Strife3dx Жыл бұрын
Shutter stock is definitely going to try and sue
@Aniketkumar-du1ne Жыл бұрын
do you use your own voice or its Text to voice generated?😂
@bomapdich Жыл бұрын
the future of porn is gonna be ... huhm
@Tod_x Жыл бұрын
it's creepy
@zengrath Жыл бұрын
Results don't look impressive tbh, seems better to just use the stable diffusion method already existing which isn't great as well. Will see what they do in later updates. Right now it's like they just generate a bunch of images from your prompt then stitch them together into a "video" that's more like a slideshow of random images with no cohesion at all. Not impressed.
@ryry9780 Жыл бұрын
I'll give it another month or two for something similar to controlnet or inpainting for videos. Still exciting to see though ngl
@oM477o Жыл бұрын
It's output looks like how text to image looked a year ago. The difference is the model has been designed for video. It doesn't look like much now but once the community works out how to fine-tune it we will probably see a huge jump in quality.
@clickpwn Жыл бұрын
@@oM477o “ Once the community works out how to fine-tune it” You mean internet perverts feeding it endless amount of porn and hentai videos right?
@MagicXStallion Жыл бұрын
@@clickpwn cant expects the devs to do all the work
@crazyaz7161 Жыл бұрын
Give the tech some time to evolve and you will be blown away in a year or two
@iamisobe Жыл бұрын
needs a different model
@grahamthomas9319 Жыл бұрын
I still think the ablitity to have more control than text prompting is the future. The human element of this stuff is the exciting bit. A machine that just generates stuff is kind of stupid. these tools work well in the hands of smart people, we are important pieces of the puzzle. God power is pointless without desire, and desire is born from experience. It is naive to asume that God power could replace the rawness that our experience generates. There is deep value in our process. If we can accept that i think it may settle alot of the fear of becoming obsolete.
@AIArtworks45 Жыл бұрын
God, i can't keep up with everything.
@kmlau1986 Жыл бұрын
Erm K, thanks but I think you shouldn't really post this in such detail down to the steps. I mean teaching people to remove watermarks, ok questionable, but teaching people how to cheat the software trial model, is probably quite illegal.
@mygamecomputer1691 Жыл бұрын
In the text I see “We are hiring” and Based in Beijing China ‘“. Is this a China product? If it is shouldn’t you be mentioning that?
@joethompson9124 Жыл бұрын
It's made in China for sure. @Aitrepreneur, I'm assuming we can convert the model to safetensors format?
@gabriel_luzz Жыл бұрын
Wating until someone manages to do NSFW videos with it rsrs
@Perry-wy2fc Жыл бұрын
C’est carrément immonde et ridicule. Faut arrêter cette merde d’IA
@EmanueleDelFio Жыл бұрын
This videos are extremely bad, i still can't figure out how this can be even compared to SD outputs , totally useless in my opinion, especially now that SD have tonz of improvements for video creations
@leemark7739 Жыл бұрын
cuda. out of memory,how to solve it
Matthew Berman
Рет қаралды 47 М.
Aitrepreneur
Рет қаралды 109 М.