Видео

i cant understand why using this ! QUESTIONS! 1. i already using Comfyui for flux , why should i use this? 2. if i want to use other models like SDXL is it better use this than comfyui ? 3. there is swarm and focuuus version , why do i use this instead of them? 4. does using this limits me in contrast to comfyui? you said doesnt support loras, is it also limited for other models? 5. since i know AI and coding whats the best option for good management and customization!

Thanks for the tutorial! I can confirm, this DOES in fact work with 8GB VRAM, I tested it with my GTX 1070 laptop, it took around 24 hours to complete but it worked.

Dont you need thermal paste? Im confused.

can you put a background thats real or is it better to use photoshop or something

Thank you very much for your Stable Diffusion tutorials. They're always great! Is it possible to install Fooocus on my local machine?

i have a question.. is the pump head also work on lga 1700 ? im gonna upgrade my 10th gen motherboard to 12gen. im using lga 1200 and i dont have the radiator box for extra pieces

in my case, i needed to clear CMOS and remove the bios batterie before the update for the process to work

Awe but the black cover popping off while u have a heart attack pulling down the lever is part of the fun!

Thank you for your hard work, its coming along fantastically, your awesome.....

ty for the tutorial, it's great to have 12GB VRAM able to train.

fooocus gives an error when I use another model, but if I use the default model: "juggernautXL_v8Rundiffusion.safetensors" then everything works without errors I select another model press "generate" and the generation hangs at the moment "Loading models ..." and then gives an error how to fix this?

Thank you. How do you update the backup bios?

Does this work on Mac M1 Chips? It looks like this might only be for Windows?

First try works perfect. Now i want to build another LoRa, but FluxGym says it's done training after 10 seconds. 😞Anybody a Workaround? (adding a num processes line in starting script dont work)

I tried training a LoRa for the first time with this, but my PC went to sleep before completing the training. It did manage to produce a LoRa model, but it's just the first 'save point'. Is there a way for me to pick up from where it left of with that LoRa? Or, at the very least, is there a saved file containing the captions for the images that I can reuse? I did them by hand and would rather not go through that again (but I would do it rather than simply using the text generator).

All im trying to do is turn my thumbnails which are pictures from game play to cartoon lol

can anybody tell me please can i use this model in comm use? and what abt style?

O'Kon Points

Has anyone actually succeeded with 8 GB VRAM?

where can i connect case cooler fan ? Any sys_fan ?

For an user that mostly cares about having as little bloatware as possible, how much do I care about not setting up windows with a microsoft account?

o te funniest thing is my husband got a good prebuild pc but if we switch te header on pump nothing works put it back on cpu fan1 it works and its a tuff x670e wifi plus d5

4060TI 16GB : 25pictures, 8 Repeat trains per image, 10 Max Train Epochs, 2000 steps. Save every 4 N epochs. 512. Result: 2 Hours.

Omg always wanted that. Thanks man 🖤🖤

So I've got a character made in SD, that I love. But can't get it to do t-poses so I can try to start making a character mesh to import into UE5. Can I use the Focus tool to keep the same character. But get the poses I want?

Why my gym is showing me some samples in training? I seeing a lot of images in process, it’s normal?

Thanks very much. Great lesson. Good tips.

Is there no option to use Vary AND InPaint together ? For example, I might be happy with the generated person, BUT would if I would like to have a variety of wristwatches, or necklaces, or rings, or whatever, to see what works best. I was hoping I could mix vary and inpaint, and just paint out the watch/ring etc, and generate a big batch of variations.

Hello, on the Civitai website, they produce incredible videos from pictures in the video tab, but it's hard to understand how they do it. Are you planning on making a few RUclips videos about them?

Hi! I was using Fluxgym for a while, it's great. But recently it stops to work. At the 66th steps it says training complete, but there is no Lora on the output file. I've retried many times but this problem persists. Is there any solution？

Thanks man. Subbed and liked.

i would have 1. used photoshop to scale people down and position them on objects, then used those images for the data set and 2. used a unique trigger word that wasn't likely to be in the base model's training set.

thanks, this is very helpfull

Fantastic video, a lot of good solid information. The lora came out great!! Thanks for the shout out!!

Hi, I have 4070, 32gb ram. Any way to get consistent Character Image? I'm using Flux.1 S with SwarmUI. Have been following your tutorial.

greate vid ur life saver

I am not expert neither, but I was thinking: maybe if you try to come up with nonrreal trigger word, because when you write "miniature person" it triggers all the neural paths responsible for plastic miniatures. Maybe if you switch the word for "m1n1ppl" or something like that, and avoid using miniature in your prompt, it might pop out better results, closer to your lora training data. But as I said, I am just guessing :D Edit: Oh. I am now just on 10:00 and I see you did that :D I should comment only after watching full video next time :D PS: You just earned a subscriber :3

Wowww BIG UPS brah!! 🎉👏🏻💯

Super helpful! Thank you!

Works good. I have a realistic renders after 5 tries. Using Mini-People and Faetastic-Details at a strength of 1.0. Would love to see a V2 of this cool Lora. Thanks! Liked and Subscribed.

This is awesome!

I downloaded this beautiful model the day it was released, and today I'm already watching your tutorial )) Beautiful work! Thank you so much!

can i train a model ?

got OOM on 3080 10gb. Did set 12gb and 512px.

I really appreciate your talk-through on this.

Some Random Tips: Let ChatGPT writ a python program for vs code that does screen captures every x seconds from a video. Tell ChatGPT that you also want to specify the start and end time and that the filename has to include the video name and the timestamp. Then you can start with some screenshots from the full video and go info details based on the result. No need for resolve an going though 100000 captures ;-)

Pretty dope!

Miniature LoRA: civitai.com/models/766577

Great tut. I have a question, How do we resume an interrupted training from the last state?