Is there way to ask you a direct questions? I am learning as much as possible about AI to help with my story narrative channel. I did buy your bundles.
I'm curious if you can use a product for a custom character. Similar to how we could train a LORA. For example, can the custom character be created using 20 images of a specific model of an iPhone or a specific model of headphones ?
Thanks for the knowledge! How can we make consistent characters in video format with them talking and having real life gestures? Is there a site that can do that?
It's hard to choose between Freepik and Leonardo AI. They both do some things better than the other. I love them both. If you want to try the Freepik Premium features free for a month, I have a code that might still work: METRICSMULE
Can you download the resulting model safetensor to be used in other platforms such as Comfy UI? Roughly what did it cost you to create the model and style?
As a paid client of your amazing offerings can you please let me know the AI-generated Influencer solution you use & possibly others. Many thanks & keep up the great work.
Its just training a LoRA. This isnt new. This has been around since Stable Diffusion 1. You've been able to do this on Hugging Face for like a year or two.
Please mention that it will eat up your 5500 credits. This ruined my entire month. Now i have to wait for the credits to reset next month to create something.
Definitely. And that's a good point you bring up. I trained the Lora with 25 images (not all realistic). When I retrained it with all realistic images using specific keywords, she looks 10x better, more realistic. I have a prompt that works incredibly well for realistic images. Let me know if you want it.
Dayum! I have a Freepik subscription but had no idea about this feature. I'll have to explore this later today.
My favorite so far for consistent characters
WHAT! WHAT! WHAT! WE HAVE BEEN WAITING FOR THIS!!! FINALLY!!!! FREAKING FINALLY DUDE! MY DUDE ERIC IT IS FINALLY HERE!
It is! And it’s awesome. 😂
This is soooo 🔝 Thanks for sharing and show how to make the most of it 🔥
Thanks for always releasing 🔥 great updates.
Thank you Eric
Very welcome
Can you have multiple characters in one scene?
Eric, everything in this video looks great. I can’t wait to try out these features! Thank you for always providing great content! 🕊️🫶
Thanks so much!
Thanks, very helpful content!
But how do you have 12 Luna images to start with? What previous tools did you use.
Great question. I used either Midjourney for character reference or Leonardo AI's character reference tools. I have videos covering both.
Is there way to ask you a direct questions? I am learning as much as possible about AI to help with my story narrative channel. I did buy your bundles.
Of course, my email is always in my description and also my profile. If needed, also 1on1 consultations, also via links in the description.
I'm curious if you can use a product for a custom character. Similar to how we could train a LORA. For example, can the custom character be created using 20 images of a specific model of an iPhone or a specific model of headphones ?
Thanks for the knowledge! How can we make consistent characters in video format with them talking and having real life gestures? Is there a site that can do that?
How many custom models does Freepik allow every month? I’m trying to compare to Openart.
Can you download the LORAs created with freepik to use locally or with other generators?
Can i download LORA after training?
Hey Eric! Did you generate your 12-25 Images for Train the Model in freepik?
Hey! Good question. Yes, I first generated 25 images first, then uploaded those to train the model.
@@metricsmule and what is your opinion? Rendernet ai or freepik? In a other video i aks you about your database. Hope you can help me.
Freepik and Leonardo AI are my #1 by far. With Freepik adding AI video like Runway and Kling now, they're going to be hard to beat.
@@metricsmule for ai Images you prefere Leonardo?
It's hard to choose between Freepik and Leonardo AI. They both do some things better than the other. I love them both. If you want to try the Freepik Premium features free for a month, I have a code that might still work: METRICSMULE
Can you do this for generating the same environments for scenes with the same characters?
Yes. If you use Flux you can use both your character and the custom styles option.
@@metricsmule I want to make only the same environment for the same scene as my character
Can you download the resulting model safetensor to be used in other platforms such as Comfy UI?
Roughly what did it cost you to create the model and style?
if you're using Comfy... just go train your LoRA on hugging face.
@@djtigon Ill have to look into that.
Any tutorials for HF?
Good point.
I’ll make one. But it’s similar to uploading many different images to train it.
@@metricsmule I used Tensor art flux model and it turned out pretty good
Can you show different male and female characters together like @Jack and @Salina Enjoy Park?
Ann you train clothes like skirts or wedding dresses?
Yes! Use the custom style for that
Dude, you didn’t even talk about the lip-synch portion of animating these beautiful looking characters
As a paid client of your amazing offerings can you please let me know the AI-generated Influencer solution you use & possibly others. Many thanks & keep up the great work.
What do you mean by solution? If it’s the creation process of my influencer, I have a video on how I made her for her music video.
@@metricsmule What I meant is did you use a company like Synthesia or any of the other Human Like Avatar solutions?
hi bro Can combine two different characters together
Hey Eric - what are you using for screen recording? Would love to know! Thank you!
For these YT videos? OBS studio. For just regular screen recordings, QuickTime or Loom.
Hey bro can i use 2 or more character at same time ? I've freepiksubscription
how do i make 12 fotos of a same caracter?
I ask the same.
great
Time to put Luna in a red dress and plug her into the Matrix
Good idea, lol
The issue is getting a dataset of consistent characters to train in freepik.
Its just training a LoRA. This isnt new. This has been around since Stable Diffusion 1. You've been able to do this on Hugging Face for like a year or two.
Yes, but it's new in Freepik, which is why Freepik is specifically mentioned in the title and thumbnail.
Grandpa and grandma not gonna watch this video. No need to spend 3 minutes talking about how to upload the images
LOL
Please mention that it will eat up your 5500 credits. This ruined my entire month. Now i have to wait for the credits to reset next month to create something.
😋😂😚
LOL 5500 credits per generation ...
Luna looks very artificial. If you create a more realistic model, will it work too?
Definitely. And that's a good point you bring up. I trained the Lora with 25 images (not all realistic). When I retrained it with all realistic images using specific keywords, she looks 10x better, more realistic. I have a prompt that works incredibly well for realistic images. Let me know if you want it.
@@metricsmule Thanks. I already created a Lora.
5500 credits one charackter, thats so cringe😂
😛😛😛 * I AM IN LOVE WITH LUNA*
😂 😂 😂