It's good you explain to some people and teach them how to use them like me I have to learn by my mistakes And I like to learn more Thank you to share this video to help us
Thank you so much for this excellent video. I accidentally stumbled upon your video, and I'm glad I did. I have been using FOOOCUS AI (based on Stable Diffusion) for about 8 months now. The biggest problem is that there is no repository of information relating to AI image generation, particularly when it comes down to understanding the obscure terminology associated with it. And the biggest issue I've encountered is trying to figure out what all the various sliders, settings, and everything else that's adjustable within the FOOOCUS GUI. As a high school teacher and college instructor, I pay homage to those who, not only understand a topic, but can explain it effectively . Thank you so much. Like and subbed.
Thanks so much for such a great and thoughtful comment! It makes us super happy that you received that much value out of it! Please come and join some of our live streams and ask any questions you have!
Thanks for this! Its so benificial to know all this as Ive just been kind of going in blind while following tutorials. Its great to have a channel willing to share so much info!
At last, a simple term introduction that tells me what a function does or is intended for in this expansive world of AI "spaghetti." Can't wait to watch the followup videos . . . keep them coming, CivitAI! --an AI noob.
Thank you very much for this content, it is extremely important for those who want to learn how to use SD, and frankly, it could be a paid course, given the didactics, amount of detail and quality of the material you are giving us for free.
Great video! It was easy to follow along and comprehend the information. I recently got a grasp of the concept of LoRAs on your website and wow what a difference it made!
ADetailer somehow managed to sneak out of this video 😄 (see 9:35, 10:14, 10:42, 10:58) Great video, nonetheless! Such a systematized and consize explanation of the core concepts is a treasure trove.
Hello! Thank you for this much valuable information. I need help with negative prompting. I remember there was an inclass word for negative prompts within the prompt. Does someone know which is it? I'm not sure if it was "--neg" or "-neg". Pls I'd appreciate the help
A link to this video is placed in the "Base Model" field in the checkpoint description on the civitai website, but I can't find an explanation here of how "Base Model" relates to the checkpoint.
I have the same question... I watched the video several times to make sure I didn't miss something but it still doesn't make sense... would be great if anyone could give a clear reply...
Even though my prompt only specifies one person, and my negative prompt specifically targets multiple people, the results keep giving me large groups of people. How do I stop that?
@@soulsborneenjoyer No, but after some diddling around and using CivitAI, I think I've sussed it out Basically a Pony checkpoint is one that's a heavily modified form of a really popular one, 'Pony Diffusion'. I think it's a case that realistic models are good for people, but you'd use specifically a Pony model if you wanted something less realistic like anime or cartoons, and using a Realistic Pony model to try and go for things like aliens, anthros, etc. with a more realistic vibe Now hopefully, someone can correct me if I'm wrong, but that's how I'm doing it!
Thanks for all the info, I'm trying to figure out more about the 'Pony' base model, since so many things on civitai is based on Pony these days. So I'm really confused when it comes to this model as there are versions of Pony based on SD 1.5, and versions based on SDXL, so when a LoRA says it's based on 'Pony', is that the SD 1.5 Pony or the SDXL Turbo based Pony? Secondly, SDXL Turbo has the same non-commercial license that SD3 has, so there could be a licensing issue there. Finally, they add their own non-commercial terms to the license, so overall, why do people still use it, when the model has quite strict restrictions on its use?
Give the article section of the website a visit and just search for 'Pony'. You'll find quite a few guides on using it properly. When a LoRA says Pony, it is USUALLY SDXL nowadays. This can be confirmed by reading the description in most cases.
@@civitai Looking further into it, Pony XL is based on SDXL, not SDXL Turbo (which has the same licensing restrictions as SD3), many sites claim it's based on Turbo, but this appears to be incorrect, which is a good thing, as I don't think we should be supporting SD3 or SDXL Turbo with its licensing restrictions. There are mixes on the official Pony page on civitai that includes SDXL Turbo, but as long as you avoid those mixes, you should be fine. Based on comments on the official discord, it seems like Pony allows you to use the outputs commercially, you just can't host a Pony based model for money, which is okay. But it would be awesome if someone can confirm my understanding here, since Pony XL is almost considered a base model at this point, and I don't think they've been very clear on what you can or can't do with the model. Regarding LoRAs, they do appear to be a mix of SDXL LoRAs and non SDXL LoRAs, so I guess as suggested by @civitai, you need to check each LoRA and figure out whether it's referring to Pony XL or Pony. Perhaps this is an improvement @civitai can make on the site, to actually have Pony XL as an option for the base model, since they are not the same thing. Though generally, when it says Pony, more often than not it means Pony XL.
Am I the only one who noticed that at 9:35 chapter 3, you have ADetailer in the index, but then you forgot to talk about it? It's not complex and almost anyone can understand without much trouble, however it strikes me that I am the only one who noticed.
Auto1111 will recognize weather the checkpoint is 1.5 or XL. You just cannot use 1.5 LoRAs or Embeddings with an XL checkpoint. The checkpoint / models themselves are interchangeable! The models are labeled which version of SD they are using on their model page :) Hope this helps!
@@civitai so it doesn't matter which model the UI uses because it will use all of them (I use easy diffusion for now) but what matters is the add ons for the models/checkpoints to be the same SD version that model/checkpoint uses. if I get that correct, thank you so much for the useful videos!
I find it insulting that CivitAi forces people to go through lenghty vids just to get some information that should be findable with simple Ctrl+F. And even after watching entire vid I still don't know how "SD 1.5-Inpainting" is different from "SD 1.5".
We're so sorry you didn't find it useful! The descriptions of the differences in Model versions are generally written by the creators in the description sections. Regardless, we will strive to do better in the future
I really appreciate your no nonsense, straight to the point explanations!
Absolutely a joy to learn this way. Really clear and I love that you overlay the text of what you are explaining so we can follow easily. Thank you
You summarized in 10min whatever im doing in the last month :) thanks !
Very helpful, thanks! Just really getting starting with generate art even though a traditional artist for 50 years.
thanks for starting the series, nice to learn something new again, I had no idea "malicious" models are a thing :)
no nonsense, straight to the point explanations!
It's good you explain to some people and teach them how to use them like me I have to learn by my mistakes And I like to learn more Thank you to share this video to help us
Thanks so much for watching! We have more videos coming up to help you out along the way. They will begin easy and work up to advanced!
you're very good at explaining things in a simple way. excellent video!
Thank you so much for this excellent video. I accidentally stumbled upon your video, and I'm glad I did. I have been using FOOOCUS AI (based on Stable Diffusion) for about 8 months now. The biggest problem is that there is no repository of information relating to AI image generation, particularly when it comes down to understanding the obscure terminology associated with it. And the biggest issue I've encountered is trying to figure out what all the various sliders, settings, and everything else that's adjustable within the FOOOCUS GUI. As a high school teacher and college instructor, I pay homage to those who, not only understand a topic, but can explain it effectively . Thank you so much. Like and subbed.
Thanks so much for such a great and thoughtful comment! It makes us super happy that you received that much value out of it! Please come and join some of our live streams and ask any questions you have!
Thanks for this! Its so benificial to know all this as Ive just been kind of going in blind while following tutorials. Its great to have a channel willing to share so much info!
Thanks for Sharing such insightful concepts. Already i m lovin with CivitAi.
Thanks so much for watching! The tutorials will start off basic and quickly get more advanced! Stay tuned!
Thank you!
Thanks for watching!
Awesome beginners Guide! Loving it! Thank you so much!
At last, a simple term introduction that tells me what a function does or is intended for in this expansive world of AI "spaghetti." Can't wait to watch the followup videos . . . keep them coming, CivitAI! --an AI noob.
Even the explanations are difficult sometimes, just keep it at!
I understand still nothing 😂
Lmao, same. 🤣
That's exactly what I was looking for! Thanx a lot!
Great introductory tutorial! Thanks!!!
Thank you very much for this content, it is extremely important for those who want to learn how to use SD, and frankly, it could be a paid course, given the didactics, amount of detail and quality of the material you are giving us for free.
AMAZING!! Thanks!!!
Thanks so much for this series!!
Dude, this is fantastic, I wish I had this when I first started learning SD
Thanks for the info.
Great video! It was easy to follow along and comprehend the information.
I recently got a grasp of the concept of LoRAs on your website and wow what a difference it made!
Thank you. It was very helpful ❤
love this series
ADetailer somehow managed to sneak out of this video 😄 (see 9:35, 10:14, 10:42, 10:58)
Great video, nonetheless! Such a systematized and consize explanation of the core concepts is a treasure trove.
Ooooops! We have to update this now anyways :)
I use "Stability Matrix" as AI image generation manager
you're a great teacher ~~>
thanks for the intro!
Hello! Thank you for this much valuable information. I need help with negative prompting. I remember there was an inclass word for negative prompts within the prompt. Does someone know which is it?
I'm not sure if it was "--neg" or "-neg". Pls I'd appreciate the help
Thanks, this helped me understand a lot of concepts🥰
It"s very helpful~
this video desperately needs an update with all the new types of models.
Great video - thanks mate
Glad you enjoyed it!!!
Thank you so much😘👍
Thanks for being here! More to come!
So Good!
amazing stuff, thanks
A link to this video is placed in the "Base Model" field in the checkpoint description on the civitai website, but I can't find an explanation here of how "Base Model" relates to the checkpoint.
who was that dancer at 3:05? guys sick! i wanna see more of his dances!
YES.
Why does the ( i ) on "Base Model Pony" link to this video, and at the "SD 1.5 / SDXL 1.0"?
Is it SD 1.5? is it SDXL 1.0? is it something else?
I have the same question... I watched the video several times to make sure I didn't miss something but it still doesn't make sense... would be great if anyone could give a clear reply...
Let me double check this for you! Sorry about any confusion!
@@civitai Base Model: Pony (i) still links to this video. The info-ball on the right side of the model type "Pony".
can i use sdxl lightning models with sd froge webui? for example i want to use juggernaut xl v9 with sd forge webui is it possible?
Even though my prompt only specifies one person, and my negative prompt specifically targets multiple people, the results keep giving me large groups of people. How do I stop that?
Do you think we will ever be able to use models on civitai, just like replicate etc.?
Question is, though, what's a Pony model? How does that differentiate from an SDXL model?
Did you find the answer?
@@soulsborneenjoyer No, but after some diddling around and using CivitAI, I think I've sussed it out
Basically a Pony checkpoint is one that's a heavily modified form of a really popular one, 'Pony Diffusion'.
I think it's a case that realistic models are good for people, but you'd use specifically a Pony model if you wanted something less realistic like anime or cartoons, and using a Realistic Pony model to try and go for things like aliens, anthros, etc. with a more realistic vibe
Now hopefully, someone can correct me if I'm wrong, but that's how I'm doing it!
hey how to add it in foocus after downloading the model ????
please what is the diff between Flux D and Flux S ? thank you so much
Thanks for all the info, I'm trying to figure out more about the 'Pony' base model, since so many things on civitai is based on Pony these days. So I'm really confused when it comes to this model as there are versions of Pony based on SD 1.5, and versions based on SDXL, so when a LoRA says it's based on 'Pony', is that the SD 1.5 Pony or the SDXL Turbo based Pony? Secondly, SDXL Turbo has the same non-commercial license that SD3 has, so there could be a licensing issue there. Finally, they add their own non-commercial terms to the license, so overall, why do people still use it, when the model has quite strict restrictions on its use?
Give the article section of the website a visit and just search for 'Pony'. You'll find quite a few guides on using it properly. When a LoRA says Pony, it is USUALLY SDXL nowadays. This can be confirmed by reading the description in most cases.
@@civitai Looking further into it, Pony XL is based on SDXL, not SDXL Turbo (which has the same licensing restrictions as SD3), many sites claim it's based on Turbo, but this appears to be incorrect, which is a good thing, as I don't think we should be supporting SD3 or SDXL Turbo with its licensing restrictions. There are mixes on the official Pony page on civitai that includes SDXL Turbo, but as long as you avoid those mixes, you should be fine.
Based on comments on the official discord, it seems like Pony allows you to use the outputs commercially, you just can't host a Pony based model for money, which is okay. But it would be awesome if someone can confirm my understanding here, since Pony XL is almost considered a base model at this point, and I don't think they've been very clear on what you can or can't do with the model.
Regarding LoRAs, they do appear to be a mix of SDXL LoRAs and non SDXL LoRAs, so I guess as suggested by @civitai, you need to check each LoRA and figure out whether it's referring to Pony XL or Pony. Perhaps this is an improvement @civitai can make on the site, to actually have Pony XL as an option for the base model, since they are not the same thing. Though generally, when it says Pony, more often than not it means Pony XL.
Am I the only one who noticed that at 9:35 chapter 3, you have ADetailer in the index, but then you forgot to talk about it?
It's not complex and almost anyone can understand without much trouble, however it strikes me that I am the only one who noticed.
Good catch! Will shoot that part and re-upload! Thanks for pointing that out!
subscribed
Hello, can i use an XL lora with a 1.0 Checkpoint model ?
No :)
how do you know which SD verison to use when downloading a checkpoint? how do I know which version my UI uses?
Auto1111 will recognize weather the checkpoint is 1.5 or XL. You just cannot use 1.5 LoRAs or Embeddings with an XL checkpoint. The checkpoint / models themselves are interchangeable! The models are labeled which version of SD they are using on their model page :)
Hope this helps!
@@civitai so it doesn't matter which model the UI uses because it will use all of them (I use easy diffusion for now) but what matters is the add ons for the models/checkpoints to be the same SD version that model/checkpoint uses.
if I get that correct, thank you so much for the useful videos!
which stable diffusion is this
There is no hope for me. This is for beginners and I’m like 🤷♂️
What's flux?
Hey I just want to have a practice session with Charlie Brown
can i sell this art adobe stock and freepik ?
No. You need to substantially change/redraw it first.
I swear to god, I heard an apple bite on 9:25
CIVIT* AI* how are you saying CIVITAIIII lol...
I know I can trust you because you're wearing a hat indoors.
No entiendo inglés.
This is not a video for begginers
英语不好真的难受。。。好痛恨自己没有好好学英语
This is too complicated, I can't understand any of this stuff.
You still spat off like a million tech lingo i have no clue and im still lost
I find it insulting that CivitAi forces people to go through lenghty vids just to get some information that should be findable with simple Ctrl+F. And even after watching entire vid I still don't know how "SD 1.5-Inpainting" is different from "SD 1.5".
We're so sorry you didn't find it useful! The descriptions of the differences in Model versions are generally written by the creators in the description sections. Regardless, we will strive to do better in the future
Thank you!