Видео

สอนดีมากๆๆ เลยค่ะ ไปดูต่างชาติอยู่ตั้งนาน ช่องคนไทยอธิบายละเอียดกว่าเยอะเลย 5555 ขอบคุณมากนะคะ

ตามมาจากเวป Thep Excel ตามดูมาตั้งแต่ webui จนมาถึงอันนี้ยอมรับว่ามึนมากๆ ผมดูวนไปวนมาหลายรอบกว่าจะวางโหนดโยงโหนดเองได้ comfy ทำให้ผมเจนภาพจาก SDXL ได้ ซึ่งwebui ทำไม่ได้เครื่องช้าไม่ก็ค้างไปเลย ผมใช้ 3070ti มันมี Vram แค่ 8GB ถ้าใช้ comfy + SDXLเวลามันโหลดเช็คพ้อยแล้วผมเช็คดูมันใช้ Vram ไปราวๆ 7GB เฉียดมากๆ แต่ webui + SDXL วิ่งไปตันที่ 8GB แล้วมันก็ช้าทันทีหรือค้างไปเลย ตอนนี้ผมเล่นเป็นได้ระดับนึง แต่ผมติดปัญหาตรงที่ว่า ผมใช้ SDXL แล้วจะอัพสเกล แล้วใช้ tile มันทำงานไม่ได้อ่ะครับ ซึ่งลองกับ SD1.5 มันใช้ได้

Thank you. This was helpful.

thanks a lot ! you're a very kind person!

Can u please upload your json workflow of this I found out how to use comfyui in android but it's hard to navigate the ui thx I'm trying to figure out how to make a github page / I figured out how to navigate comfyui u can zoom out and link nodes just tap and drag😊

thanks!

This is gold standard....many vidoes in one video actually.....again a great great video....but sir what is your hardware configuration for generation ?

This is superb !! I want to give you 10 likes actually but RUclips limites it to one. Keep up good work. Sir, Can you let me know your hardware specification for this type of work ?

That was amazing!!! BIGLOVE!!

ทำแบบนี้ใน sd ได้ไหมครับ

thank you so much

It realy helps! Thanks.

Excellent - thanks

มีพากย์ไทยที่ไม่ต้องอ่านซับไตเติ้ลไหมครับ?

you sound soy

Can you do a video to make an image clearer/sharper more detailed. Sometimes I get an image and I like the general look and poses but I am not able to get a high quality version of it without major weird changes.

กด queue Prompt อย่างไรครับ ให้มันทำแค่ ksample เดียว ผมกดแล้วมันทำงาน 2 อันเลยครับ

Thank you

Great material, thanks a lot!

I was watching Latent Vision but the way he teaches things is very advanced. You really helped me understand things from the very basics.

You’re a great teacher! So sad you haven’t done more videos in the last 6 months😢 I’ll keep subscribed and hope you do more. Thank so much for what you’ve already done!

I've been playing with Reactor and now trying to add detail to the smooth airbrushed faces. This series is really helping me, though the face restore models are driving me nuts. One adds skin detail but wonky eyes, the other has good eyes but less skin pores. Anyway, on with the series, see if you introduce better ways :). Your tutorials have really explained some basic core concepts of how AI works that too many others skim through without explanation. Really appreciate that.

is there any way i could use the "confusing" tradicional comfyui method to do a lora stack\multiple loras? I don't like the efficiency method 💀

Good!

ไม่ทราบสเปกคอมที่ใช้เป็นแบบไหนครับ เจนเร็วมาก

Did this video REALLY need to be 22 minutes long?

Love your Explaination, I'm a new bies.. detail n understand enough. Thnx

Why do my images turn out opaque? :(

hello, is it posible 2d character to psd file, each body parts another image file, how can we do?

How can I turn on progress image on ksampler node(box)?

efficiency มันบอกของผมต้องใช้ comfyui version 2.0 ++

เป็นคลิปที่ดีมากๆ ฝึกตั้งแต่ง่ายๆ ไปยากขึ้นเรื่อยๆ ผมก็นั่งทำซ้ำๆ จนจำได้แล้ว ขอบคุณครับ

เพิ่งหัดเล่น เข้าใจง่ายดี พี่ทำคลิปเยอะๆครับ

Oh my God. This is the easiest basic comfyui tutorial channel for me to understand. I was able to understand each step and how to connect the nodes together. Thank you.

Thank you for showing the several methods of masking and inpainting!

เข้าใจง่ายดีครับ

Does he go over how to use reference image?

In the first example, what if I want to do an image to image using controlnet, like to recreate the same or similar image but with a different pose, how would i do that, pls help

Absolutely awesome video! Thank you!

Great tutorial, I was struggling trying to understand how ControlNet works but now is very clear. Thanks for sharing your knowledge!

Thank you for great great tutorial.

This video is from over 5 centuries into the future. Time travel is real, people!

I know the video is old, but maybe you know. I just discovered thanks to you the efficient nodes and I 'm trying them a bit when I found a problem. With UltriumV7SDXL model, the efficient pipelie gives a pattern image that looks like noise and the standard comfyui does give a proper result. Do you know if it's a known issue or if I am doing something wrong? Btw, good video! I suscribed :)

Спасибо бро

Amazing video, very clear

New person question: When you have all of this hooked up to your main workflow and you hit "Queue Prompt" it runs through the whole thing instead of going to the mask subroutine, screwing everything up. I didn't see you bypass anything so how do you avoid that?

Thanks mate! Easy way to setup lora!

Hello, when I add the image in order to get the nodes that were used to generate that image it is not working for me, maybe I am doing something wrong?

Sometimes inpainting through SetLatentNoiseMask generates very poorly compared to using Vae for Inpainting, poorly meaning that though it stays in the mask, what it generates is not logically good or desired compared to Vae for Inpainting which seems to 'understand' better what is intended or desired. Have you seen this or have advice? in your example at 6:51 using SetLatentNoiseMask works great so i dont know why i sometimes get poor results unless it is the model i am using. do some models work poorly with inpainting and can you provide your experience on that? Thank you for your channel

คนไทยใช้มะ