Thanks, Man!! Awesome stuff. I use MiniMax every day and used it to make a quick teaser trailer for my AI web series that's premiering next month. It's dope what you can create with it, as it is right now much more advanced with following prompts than the others.
Thanks! Minimax works really well, sometimes it just takes a few subtle adjustments to the prompts to get something that looks like what you want. But the Ai is capable of creating all sorts of different motions and styles
You explained a lot more detail than most on this generator. I didn't know they were censoring! Thanks for the excellent prompting tips. I can hardly wait until the image to video. Others have said it does a better job when translating to Chinese then putting that prompt in. Yours seem to kick ass in English. I'll try both.
I got you covered 👍. English prompts seemed to work fine for me, I haven't noticed much of a difference in changing the language I'm prompting in overall. Image to video is going to be insane!
Ai will often add in some camera motion to the video. But if you want more control like a "director" it's good to use your own camera motion in the prompt
I've found that Minimax does a pretty good job of following simple camera motions, sometimes when you try to combine multiple motions it gets confused though
Hey Tao, thanks for the video! The problem I have with MiniMax is that you cannot do consistent characters. Also, MiniMax videos just "end" they do not allow for extending. I tried a work around by taking the last frame of a MiniMax Video and using it as an image upload in Luma for image to video. This trick works but is not that great in quality. I think its fun to create a lot of free 5 sec vids but for serious generations, I'll wait for the next upgrade from MiniMax.
I've heard that they are planning to release an image to video👍, it's really good especially since it's free and apparently there's only a team of 6 people working on it
Thanks for the video. Sometimes, it doesn't follow the instruction set in the prompt. For cinematic shot, it sometimes produces disney pixar style. It is frustrating. Do you know the workaround?
i made 1 similar to the one above changing the setting to a ruined city and mutated the creature turned out pretty good I think, you have new sub thanks Tao
Absolutely, it's all in the prompt. I did a few nice stories already and so far quite consistent, especially with animals. Will do some sci-if ones. Yes, I multi-task and have all the scenes I want in mind, and do it one at a time..sometimes over a few days.. but worth it.
Hey, love your content! What LLM do you recommend for generating the best image or video prompts? Chatgpt or Claude? I’m finding that not all of them are following prompt formats.
Great vid, I am just wondering how do you keep building on the same prompt and characters, every time I try to create it is a new character from the last it is never the same a new character is created. Is there a way to build on the previous prompt, so you could make a longer format video that has consistent characters. If I missed this please direct me. Or is it only way is by using a photo reference. Thanks so much!
Photo reference is the best way. You could be super descriptive in the prompt: facial features (eyes, nose, hair), exact clothing and colors. However based, on what I've seen that works best for pixar style characters, not realistic ones.
There's not really anyway to do that right now. The closest you can get is to prompt for the same character archetype, appearance details and visual style
Hey Tao ! your video is awesome, i've watched many of them, but i found one important thing that you're missing is camera type and director, which are very crucial in film making. Can you do an advanced video listing about that ? i need that so bad 🥲
Thanks for the suggestion! I typically find that using image references is the best way to get Ai videos to look the way you want. I made a guide for camera film types here: ua-cam.com/video/Cv0J6fGS5Cg/v-deo.html
Closeup shot of a hybrid female humanoid plant creature. The hybrid humanoid creature has a sad expression. Cinematic movie. Muted and desaturated color grading, post-apocalyptic movie scene directed by Andrei Tarkovsky
After finding this video, I watched it and it was too good to be true. I had fun creating random creatures and hybrids tonight. I hope they’ll add image to video soon, that will be next level. But your video was awesome
When viewing other videos that people have created it shows part of the prompt they used. Is there a way to see their full prompt? The only option it gives me is to download it but I can’t see the prompt used if I want to use it as a base prompt to mine. Hope that makes sense
Hi Brother, This currently creates videos only for 5 seconds and I can't see another option to extend the length. Is there something that passed by your sight regards to this?
Hi, I keep trying to use it, but it goes into a chat instead. Am I doing something wrong? Im bypassing the inital autentication park, can that be affecting it?
Hello tao, love your video, please can you make a video on creating multiple consistent characters on midjourney website, because i tried the prompts you showed in your previous video, but it didnt work, thank you.
Klingai does the same for censoring, I found that the words "dirty", "chest", "jagged" are censored, I had found 7-8 more words, I will share once I find that text file :)
I wonder if they are censoring the text itself, or censoring the video after its generated if they detect "bad" content in there. I wish they'd just tell you what's censored though
Tao Prompts I need your help please. Everyone with ai channels just seem to be clueless about what I want. Ok here is my issue I have a picture. I want AI to make the transform the picture into a talking video. Only with head movement and a few hand gestures. I just want a simple method to upload my pic and video and let the ai do the job of converting the picture to the video. I tried replicate (yoyo something like that) but it keeps giving me an error. I dont mind paying for replicate. But even the pricing I dont even know what to do because most of these ai sites are so complex and confusing as a beginner. I'm new to AI. Any ideas on how to do this?
@@RobertsDigital here is my latest tutorial, there's a few different tools used here but you can generate consistent characters doing different things: ua-cam.com/video/ltuRxvaCdMs/v-deo.html
Hey man can you do a tutorial on how you think the dark fantasy video tiktoks are made through luma lab dream machine or through other video editing toools?
@Bartetmedia It's not a problem because it is easy to remove the watermark, and for free, too. Or, in some cases just enlarging the video when editing the clips together. Image to video is about to come to MiniMax, though, but MiniMax as it is now is still far superior in accurately following prompts and in creating accurate human movements than the other image to video programs out there. Kling doesn't even come close to MiniMax in following prompts. I've tested this.
Oh bro, your videos are far better than paid courses honestly... Thanks (y)
Thanks Ali, I try to make the best tutorials I can 👍
@@taopromptsFree or pay ??
best ai channel out there. you deserve a lot man! keep up the vibe ❤
I appreciate that man, Ai video has had so many exciting updates recently 👍
Thanks, Man!! Awesome stuff. I use MiniMax every day and used it to make a quick teaser trailer for my AI web series that's premiering next month. It's dope what you can create with it, as it is right now much more advanced with following prompts than the others.
Thanks! Minimax works really well, sometimes it just takes a few subtle adjustments to the prompts to get something that looks like what you want. But the Ai is capable of creating all sorts of different motions and styles
Tao, this is amazing. Thank you and will check it out next week!!
I hope you tried Minimax out, it does an great job of following your prompts 👍
@@taoprompts Really? Okay, will try it out! Thank you Tao!!
You explained a lot more detail than most on this generator. I didn't know they were censoring! Thanks for the excellent prompting tips. I can hardly wait until the image to video. Others have said it does a better job when translating to Chinese then putting that prompt in. Yours seem to kick ass in English. I'll try both.
I got you covered 👍. English prompts seemed to work fine for me, I haven't noticed much of a difference in changing the language I'm prompting in overall. Image to video is going to be insane!
Great review !
The potential of minimax is insane
It's next level. I've tested it out for a few of my animations
Check out the Cat heist Ai Music Video they made with #minimax also.
No links so you'll have to search it on YT
I'm going use it too now
Agreed, the text-to-video generations are better than other Ai video platforms I've seen so far
@@taoprompts Totally. have you seen Edski Studios music video generations?
WOW. I just saw a few examples of minimax and came straight here to see if you had a tutorial. So glad you're on top of it sir! Well done!
Minimax can do some amazing stuff if you have the right prompts, now it just needs Image-to-Video!
Very useful tips I always learned a lot from your video you're the best thank you for sharing.
Thanks man, it looks like Minimax is pretty amazing at text-to-video. Now we wait for image-to-video 🙏
@@taoprompts who makes this AI videos
Best channel for ai video generation prompt tutorial👍
Thanks! I'll keep you updated on any big new releases
Playing around with MiniMax and so far it's the best out there.
For text-to-video it can generate a ton of motion compared to other Ai's
So awesome, will dig into today Tao bro!
I hope you got to play around with it, it's one of the best text-to-video imo
Nice vid Tao, I’ll check out the CapCut video upscaler as well
The capcut video upscaler works pretty well for realistic videos I've found
at 3:51 you did not add any camera shot+motion. But the ai still gave you something good. Is it optional to add the camera shot+motion?
Ai will often add in some camera motion to the video. But if you want more control like a "director" it's good to use your own camera motion in the prompt
Crazy stuff man. Crazy stuff
Ai video is going crazy man, it kind of reminds of when Midjourney V6 first came out and the quality was so good for that time
how great!! I am so lucky that I find of your UA-cam!!
I'm glad you found me! I've got some more guides for Minimax soon 👍
great tips.. used the tool last week, didn't realize i could get better outputs and the camera angles; great tip
I've found that Minimax does a pretty good job of following simple camera motions, sometimes when you try to combine multiple motions it gets confused though
Thank you! More of this please
Of course! I plan on more minimax guides like this
This is tremendously helpful you have opened my eyes and my mind to what is possible.
Great to hear that man, it feels like Ai video advances a crazy amount in such a short time
keep up the great work bro! thank you!
Thanks man, I've got a lot more videos planned!
Very nice.
I've used all the ai video generators.and over all I think minimax is pretty dam hot, and its free.
It's free and also pretty fast, I haven't had any issues with queue time at all!
Hey Tao, thanks for the video! The problem I have with MiniMax is that you cannot do consistent characters. Also, MiniMax videos just "end" they do not allow for extending. I tried a work around by taking the last frame of a MiniMax Video and using it as an image upload in Luma for image to video. This trick works but is not that great in quality. I think its fun to create a lot of free 5 sec vids but for serious generations, I'll wait for the next upgrade from MiniMax.
I've heard that they are planning to release an image to video👍, it's really good especially since it's free and apparently there's only a team of 6 people working on it
Thanks for the video. How do you stop the characters from moving their lips. It seems they are talking in most video generated with this ai video tool
It helps to use words like, "static" or "subtle motion". But it does like to animate people speaking, sometimes it just takes a few tries
Thanks for the video. Sometimes, it doesn't follow the instruction set in the prompt. For cinematic shot, it sometimes produces disney pixar style. It is frustrating. Do you know the workaround?
You can prompt for the color palette like "muted colors" "desaturated colors", also a specific film type or genre
i made 1 similar to the one above changing the setting to a ruined city and mutated the creature turned out pretty good I think, you have new sub thanks Tao
That's great! Good to know this also worked for you. And thanks for subbing
@@taoprompts your very welcome♥
What about scenes with dialogues(mouth movement) ? Did you try it and were they acceptable in terms of consistency?
Minimax doesn't have lip sync right now. But speaking motions work well
Thx for the video, great clips!
Absolutely, it's all in the prompt. I did a few nice stories already and so far quite consistent, especially with animals. Will do some sci-if ones. Yes, I multi-task and have all the scenes I want in mind, and do it one at a time..sometimes over a few days.. but worth it.
For sure, I've seen people put together some pretty impressive videos with Minimax, just with text-to-video
thank you , I use my custom GPT, and generate all my prompts , results, amazing
That sounds like it would help a lot, I need to spend some more time checking out different prompt GPTs
Hey, love your content! What LLM do you recommend for generating the best image or video prompts? Chatgpt or Claude? I’m finding that not all of them are following prompt formats.
I typically write the prompts myself. I've used ChatGPT before, I haven't tried Claude yet.
Great vid, I am just wondering how do you keep building on the same prompt and characters, every time I try to create it is a new character from the last it is never the same a new character is created. Is there a way to build on the previous prompt, so you could make a longer format video that has consistent characters. If I missed this please direct me. Or is it only way is by using a photo reference. Thanks so much!
Photo reference is the best way.
You could be super descriptive in the prompt: facial features (eyes, nose, hair), exact clothing and colors. However based, on what I've seen that works best for pixar style characters, not realistic ones.
@@taoprompts thank you! You the man!!
Nice and informative. Thanks
thank you for your hard work
Thanks, I appreciate that!
Thanks for the information
Is it possible to obtain a 9:16 video format? I'm trying but no luck
I don't think you can do that right now
The day they release video to video option it will be a game changer
For sure. In the meantime though, Runway has a decent video to video Ai
@@taoprompts sure.. the problem is that its quite pricey :/ hope they drop the price a bit in the future
Sir you always make best and unique video with all details thank you so much
Thanks I appreciate that, I hope these tutorials are helping you out 👍
How to prompt get video out as 9:16 aspet ratio?
I don't think that's possible right now
@@taoprompts ok
Great! Can I adjust the scale to 16:9 ?
I believe it's 16:9 by default
is there a limit in the free generation or daily limit?
I did not run into any limits when using it
Will these videos be monetized?
Are you making a Kling 1.5 video? I don't get the motion brush.
motion brush only works for version 1.0 atm
Excellent content. Very helpful indeed.
Thanks Allan!
Hi Tao. What's the maximum length of videos that can be generated?
For minimax they are 5 seconds
how can we keep the character consistency using minimax?
There's not really anyway to do that right now. The closest you can get is to prompt for the same character archetype, appearance details and visual style
@@taoprompts thanks
How to I create music video using ai
My music video using MiniMax:
ua-cam.com/video/P4s90CBrgOw/v-deo.htmlsi=tBOh35moL_GXA3rG
@@JurajPecháč I mean using my own face
@@rajdebbarma88 Hedra AI:
ua-cam.com/video/RPrDU_vPtCI/v-deo.htmlsi=qC4i4R9TgP1Rh5dL
I made a tutorial here on how to put yourself into an Ai video ua-cam.com/video/8UDJERXlPmg/v-deo.html
@@taoprompts ok
How you create background sound
I use stock music from storyblocks
Thank you amazing 😀
How do you extend the video from 5 seconds? Luma AI has it. Is there a such feature on King AI?
Yes you can extend 5 seconds in Kling AI. For Minimax I don't' think that's possible
@@taoprompts so it does extend but only with kling 1.0 it doesn’t extend with 1.5.
Great video! Do you know where I can find a detailed explanation on text of how to prompt in this AI?
anakin.ai/blog/minimax-prompt-masterclass-unlocking-the-full-potential-of-ai-video-generation/
❤ great video! Music is fantastic too, where do you get that from?
I use stock music sites like storyblocks or artlist
Very impressionnant ❤❤
Thank you! Minimax does such a good job of following prompts
Hey Tao ! your video is awesome, i've watched many of them, but i found one important thing that you're missing is camera type and director, which are very crucial in film making. Can you do an advanced video listing about that ? i need that so bad 🥲
Thanks for the suggestion! I typically find that using image references is the best way to get Ai videos to look the way you want. I made a guide for camera film types here: ua-cam.com/video/Cv0J6fGS5Cg/v-deo.html
cool. thanks very much. gives me some ideas how to play with this.
Minimax is pretty amazing, I can't wait til they release the image to video
Another Excellent Video, I was having the same problem with the Generations, I would bring the percentage up, then completes then dissapears
Thank you, that issue is pretty annoying. Definitely something they'll need to change imo
Thanks for the video! Can we get the prompt of the lady with veggies on her head?
Closeup shot of a hybrid female humanoid plant creature. The hybrid humanoid creature has a sad expression. Cinematic movie. Muted and desaturated color grading, post-apocalyptic movie scene directed by Andrei Tarkovsky
@@taoprompts thanks Tao! Is this for the lady you have in the thumbnail?
Thank you for this. Very clear and informative video. I subscribe to the channel.
Glad you liked this guide, and thanks for subbing!
Great video 💥💥💥
Thanks Sean!
@@taoprompts Dear Tao, Please make a video about characters talking. How to create characters with flawless lip-sync?
Can i use minimax generated ai story videos on my monetize UA-cam channel
it is not working now, i don't know
Super nice video and channel! Is it possible we don’t get the code to login in Switzerland ? Never got the code despite trying many times…
Thank you! You might be able to use it without logging in. Just try clicking the second green square and prompt for something
@@taoprompts yep, got it! Thanks for answering 🙏🏼
Please, will i have issues with copyright if i post videos with the watermark?
I'm not sure, but my guess is if you leave the watermark there it's probably fine. I doubt anyone will care.
Nice bro
After finding this video, I watched it and it was too good to be true. I had fun creating random creatures and hybrids tonight. I hope they’ll add image to video soon, that will be next level. But your video was awesome
Minimax can make amazing videos, sometimes it takes a few tries and adjustments to the prompts. But all these videos are definitely possible
When viewing other videos that people have created it shows part of the prompt they used. Is there a way to see their full prompt? The only option it gives me is to download it but I can’t see the prompt used if I want to use it as a base prompt to mine. Hope that makes sense
If you click on another persons video it shows the full prompt on the right side
Thank you 😊
why do you always nee to add "directed by andrei tarkovski" ?? Does the watermark gets removed on the paid plan?
watermark is removed with paid plan. I add in the director to try and keep the visual style consistent
Hi Brother,
This currently creates videos only for 5 seconds and I can't see another option to extend the length. Is there something that passed by your sight regards to this?
Right now, it's only 5 second videos
They have limited videos per day?
I never got limited when using it
Hi, I keep trying to use it, but it goes into a chat instead. Am I doing something wrong? Im bypassing the inital autentication park, can that be affecting it?
It looks like they updated the interface a bit, you now can use an email account to sign in. Make sure you click the second green rectangle
Hmm when I click it get site with Chinese text. How to select English option?
it should have an english website here: hailuoai.video/
Pls do you know any image to video Ai generator that’s good that’s not runway
Kling Ai or Luma Ai are really good
Hello tao, love your video, please can you make a video on creating multiple consistent characters on midjourney website, because i tried the prompts you showed in your previous video, but it didnt work, thank you.
I'll work on a new guide for MJ version 7 👍
Can I use my own photo in the video
This doesn't have image-to-video yet
@@taoprompts ok thanks so much for answering, I appreciate it
Klingai does the same for censoring, I found that the words "dirty", "chest", "jagged" are censored, I had found 7-8 more words, I will share once I find that text file :)
I wonder if they are censoring the text itself, or censoring the video after its generated if they detect "bad" content in there. I wish they'd just tell you what's censored though
super! супер!
Tao Prompts I need your help please. Everyone with ai channels just seem to be clueless about what I want.
Ok here is my issue I have a picture. I want AI to make the transform the picture into a talking video. Only with head movement and a few hand gestures. I just want a simple method to upload my pic and video and let the ai do the job of converting the picture to the video. I tried replicate (yoyo something like that) but it keeps giving me an error. I dont mind paying for replicate. But even the pricing I dont even know what to do because most of these ai sites are so complex and confusing as a beginner. I'm new to AI.
Any ideas on how to do this?
Kling ai does a good job with this, if you prompt for "the man is speaking, [whatever hand gesture you want]", it works pretty well for me
@@taoprompts with a consistent character?
@@RobertsDigital here is my latest tutorial, there's a few different tools used here but you can generate consistent characters doing different things:
ua-cam.com/video/ltuRxvaCdMs/v-deo.html
@@taoprompts Alright I'll have a look at it.
Thanks.
Love it
1st 😘😊
Let's go! 🔥
Hey man can you do a tutorial on how you think the dark fantasy video tiktoks are made through luma lab dream machine or through other video editing toools?
I haven't seen those before, do you have an example video link
hi cool video)) tell me how can I remove a watermark from a video? Is there a neuron for this?
HitPaw has a watermark remover
@@taoprompts thks bro))
Bro how to I add my face on this scene, make a tutorial
ua-cam.com/video/8UDJERXlPmg/v-deo.html
You can use a video watermark remover😊
The problem with Free is that the clips are produced with a watermark otherwise it's pretty good. I prefer image to video.
I've heard that they plan to release image to video soon
@Bartetmedia It's not a problem because it is easy to remove the watermark, and for free, too. Or, in some cases just enlarging the video when editing the clips together. Image to video is about to come to MiniMax, though, but MiniMax as it is now is still far superior in accurately following prompts and in creating accurate human movements than the other image to video programs out there. Kling doesn't even come close to MiniMax in following prompts. I've tested this.
Hi. Why its runway that other tools do not introduce the option - character to video? How vidu?
You can crop them out the watermarks
@@taoprompts next level omg
this is dope! but i noticed this doesnt allow image to video
Right now it's only text to video, I've heard they plan to release image to video soon though
@@taoprompts hope so cuz they are pretty good
The main problem is consistent character, how to constant same character on every scenes?
There's no good way to do that right now. You can try to be specific in the prompt about appearance and character archetype
How can i create a perfect prompt with chatGPT, you can any help ?
I don't use chat gpt to create prompts, I don't know how that would work.
I've used this app on my phone, but when I try to log in, only China phone numbers are accepted. How can I solve this issue ?
does the minimax website trigger anyone else's computer security ? I get a malicious threat warning, can't access the site
Good job,tried decent for free account.
Runways gen 3 turbo hai vertical videos launched please make video on that
Thanks for letting me know!
❤
Thanks for the love!
Bro do you have a discord channel ?
I don't have a discord channel atm, maybe in the future
Send link pls😊
It's in the description
Hello my friend,
Mention please that Minimax does not allow commercial use in your next video about them
You don't need to login, just click the second green create icon below
Thanks for the tip 👍
thanks for ur video but it takes a long time to generate a video
Yeah, it's free right now so it has a huge backlog of users
this is crazy
100%, it's a level above everything else
Do you actually use any of these Ai video generators for your business or work?
Yes, I work with various artists to create Ai videos for them. Professionally I work as a researcher in the AI industry.
I like the video, but I can't login on their website, either by phone or qr code, anyone else having this issue?
You might be able to generate videos without logging in, just directly try clicking the second green box and prompt for a video
Check out the new kling update..
i give you one tip, USE THE LENS BRAND, not the CAMERA; so it wont show the CAMERA
Thanks for the suggestion, that does work better
@@taoprompts yes. I just upload new music video. Minimax ...
@@gumvue.studio It looks great! A 4-5 minute video like you made takes a lot of work 👍👍
@@taoprompts yes, but I use my custom GPTs, that cut the time 10x