Ride My PONY - REAL Edition
By pAInCREAT0R
Time to ride the PONY
This is the most realistic of my three PONY models. I am making it available for a limited time because of numerous requests. It was trained on a similar high definition image set as my SDXL realistic models, but it retains the ability to do the stuff that PONY models are known far. I still prefer SDXL for most applications (or FLUX), but the prompt adherence of PONY is always impressive.
V4: Trained on more faces and photo quality pictures. Leans more to female renders but will still do both with proper prompting. Less artsy though than V3 if that is what you prefer. I will update comparisons below soon for reference. NOTE: Eyes do better if you mention them in your prompt!
V3: Added a new trained photoset and improved block merging to smooth results backwards. Existing PONY LoRA's work on it. Character LoRA's trained on it shine though. Here is an example of the difference between versions:
This is NOT a lightning or hyper model. It performs best at 25-55 steps. CFG of 4 has worked best for me.
Disclaimer: This model is provided for educational purposes and for the advancement of AI art in general. The creator of this checkpoint assumes no responsibility or liability for the end user's outputs. Any illegal, unethical, or deep fake use of this model is the sole responsibility of the user engaged in those activities. The end user assumes all liabilities from use of the checkpoint.
No permissions are given for anyone to distribute this model on other sites. I do not give permission for it to be used on any image generation service other than CIVITAI.
Description
V4 went through a lot more training focused on better faces and eyes. The results are good, but it did introduce a stronger bias to female characters. It will lean strongly toward women unless prompted otherwise. V3 is more flexible in this regard if you need more versatility.
FAQ
Comments (9)
Thank you, I love all of your models.
Very good model, but i think anything over 30 steps is non-vital, just a little more detail, at least on 2m Karras.
So its safe to use at 30 steps for those, who dont have a 4090:)
How do yall make those animated images? Whenever I do it, it's just a still image that zooms in...
thanks for your awesome models! <3 sunspin99 :)
Finally a model that can make a snowman with his carrot nose in between Ms. Santa Claus's butt cheeks. Cheers & thanks you legend!
遵从度很高
Best since V1 with much more realism. What a week. It can play on various sampler scheduler combo. With max 4 cfg you can use the fanciest samplers, for example 3msde, because it doesnt tend to anime. It needs more steps, than usual, 35-40.
your models are so cool! thanks for crafting them. :) xoxo sunspin99
thanks for this one!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.