Trigger tina2
Steps 20
Sampling SDE KArras
Use Adetailer
Description
FAQ
Comments (9)
you make good loras but I think you should try DPO in XL I am going to leave you some of the information I have been researching these days
(copy paste)
here is a list of all the models that I have seen and tested that have DPO
1-AutismMix SDXL
https://civitai.com/models/288584?modelVersionId=324619 (all three of their model mixes have DPO)
2-Confetti Comrade Mix
https://civitai.com/models/280344/confetti-comrade-mix(here I had left some comments giving ideas to move the community forward, anyway I will leave my ideas in this post in case someone reads it and wants to know more about it)
3-Js2Prony
https://civitai.com/models/280321/js2prony
4-Comradeship XL
https://civitai.com/models/246299/comradeship-xl(my favorite so far)
5-[0.8(ponyDiffusionV6XL_v6) + 0.2(reproginexl_v10)] > NAI 3 !
https://civitai.com/models/263797/08ponydiffusionv6xlv6-02reproginexlv10-greater-nai-3
6-Hassaku XL
https://civitai.com/models/140272/hassaku-xl-sfwandnsfw-beta(this model understands prompts well but is not good at replicating characters without loras,the other models in the list including the Pony Diffusion V6 XL can replicate characters perfectly,even better than Loras already made in the civitai community)
the list I just left have good implementation of (DPO) the idea of this post is that the creators already famous in civitai begin to implement this technology,98% of civitai models are mediocre but with this technology the existing models can make magic,genitals,legs,hands,expressions,poses, concepts etc make more sense with these models already named(Pony Diffusion V6 XL was the first to implement the technology to a XL model) but the ones already named here are of very good quality also it would be great to continue seeing models with many different styles and more polished in the future) researching I discovered that if you mix a model that already has(DPO) with another model already existing civitai gives results like these. ..the technology works very well with XL but as far as I can see very bad with 1.5(for the moment) I think the future this year is XL I think the other model and lora creators should all work for XL,it feels good when you put indications and you get almost everything as you want,that with 1.5 is impossible,I am using XL,without controlnet and without Lora I have had incredible results,the mixes with the DPO models I think it is better to do them in Comfyui,I hope this post can help you,there are many people who still don't know about the technology
I took the time to test about 50 XL models and all were exactly the same as the 1.5,bad anatomy etc,in search of DPO models I did my tests by replicating famous characters,an effective way to test is using as a prompt:
cammy,chun li,jinx,league of legends,etc,if characters like that replicate perfectly the model has DPO,if the replica of the character doesn't come out then it doesn't,the models in this list can replicate the characters already named and many many more,(just do your list of character tests)....
PS:there are models in civitai that claim to have DPO and don't have them or are very bad,all of them were tested by me so I save you time
here you can read the post of this page to know a little more about DPO
https://civitai.com/models/242825/dpo-direct-preference-optimization-lora-for-xl-and-15-openrail
Wow, I had no idea about this, I'll approve it but is it necessary to update to XL for good results or does it work well in 1.5?
@Marlosart for now there are very bad results in 1.5,the structure of XL allows much better results and understanding of prompt,if you read all the information I left here you might find something useful and learn how to do the loras,I think 1. 5 has been surpassed since DPO technology was implemented,you can do your own tests and compare,you can start by downloading one of the models I mentioned and just with prompts you will realize how flexible and good this is,there are a few style loras with DPO and a few character loras,the current models with DPO are already very good replicating to perfection hundreds or thousands of already popular characters,just put as prompt for example: characters from overwartch,league of legends,etc,not all existing characters are perfectly integrated in the DPO models,but there are many,the best is to go creating loras that are DPO for these models,you can go guiding you of the loras of the community,with write in the search engine(for pony) appear some loras already compatible with DPO,try to see if it works for you and create the paladins loras with DPO that the results with the models with those characters are not so good.
@FireBlack85 I have an other question? Can I have the 1.5 and the xl at the same time? If you know how it would help me.
@Marlosart I don't understand very well the question,but if you are interested in creating your Loras for pony I can help you,I have created a few that I haven't published in civitai because I am still testing parameters and how to make them more consistent,maybe with your experience creating Loras you can help me to improve,if you use Khoya SS to train this is my configuration https://mega.nz/file/6RV3xbhS#600N3w_wnsZCH6kGdhCVkyuiUC9gxo1J9XKWO1a5EuI
I have been testing with 16,20 and 30 images in training having regular results,for some reason I don't make my loras versatile by changing the style from one model to another.
here is a Maeve I made a few days ago
https://mega.nz/file/GVlWTTrL#QJlkPyTaeJuAZf7KJHHIsYaE0XhckvU324kCZYtbTxc is compatible with pony
you don't have to make all your dataset from 0,just mix pony with one of your 1.5 dataset you have made and that should be enough,the trigger word of the maeve is the same name of the file,if you have any questions let me know
@FireBlack85 I was wondering if it is not necessary to delete 1.5 to install XL
@Marlosart you don't need to uninstall 1.5 to install XL, just download the XL model you want to try and that's it, 1.5 and XL work in the same interface, you can use both in automatic1111 or comfyui,if you are going to try it download pony xl and try what i told you
@FireBlack85 I have a problem, my PC does not support the XL models. Every time I try to create an image, I get a cuda error.
I love this Tina's lora. Do you have more DOA girls?