---
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
widget:
- text: >-
    score_9, score_8_up, score_7_up, BREAK, innocent and alluring, long hair,
    alternative vibe, beautiful eyes, freckles, medium breasts, subtle cleavage,
    tight body, slutty, candid picture, (cozy bedroom), off-shoulder shirt,
    dynamic angle, vibrant lighting, high contrast, dramatic shadows, highly
    detailed, detailed skin, depth of field, film grain
  parameters:
    negative_prompt: score_1, score_2, score_3, text
  output:
    url: images/00000-253699598.jpeg
- text: >-
    score_9, score_8_up, score_7_up, BREAK, woman, bikini, skin pores, little
    smirk, sunglasses, pool, sunny, 80s vibe, depth of field, film grain
  parameters:
    negative_prompt: score 1, score 2, score 3, text, penetration
  output:
    url: images/00021-3188932120.jpeg
base_model: LyliaEngine/Pony_Diffusion_V6_XL
instance_prompt: None
license: cdla-permissive-2.0
---
# ponyRealism_v21MainVAE

<Gallery />

## Model description 

⚠️ Please read the description

🏋️‍♀️ For training LoRas using the model go here


♻️ v2.1 Is a major rework of the model.

    Main version has the most detail at the cost of Sampler compatibility

    Alternative version works on most Samplers at the cost of some detail

ℹ️ This version leans more towards Pony for more overall variety, but maintains the realistic quality.
ℹ️ Example images are without Adetailer


📍Recommended Parameters For the MAIN version:

· Samplers:

    DPM++ SDE

    DPM++ 2S a Karras

    DPM++ SDE Karras

    DPM++ 2M SDE Karras

    DPM++ 2M SDE Exponential

    DPM2 a

    DPM++ 2S a

    DPM++ 3M *

I DO NOT recommend the use of the following samplers (weak detail):

    Euler a

    Euler

· Steps: 20 - 30 .

· CFG: 6-7

· Clip Skip: 2
· Resolution: I have been using 1280x768 | 768x1280

· I recommend using female&#x2F;male instead of woman&#x2F;man

ℹ️ Samplers not listed on the recommended won&#39;t work or need some tweaking in CFG or steps.

This version does not need any trigger words, and as like most Pony models it uses the following prompt style:

Positive:

score_9, score_8_up, score_7_up, BREAK

Negative:

score_4, score_5, score_6

💪The model has more general variety, creativity and better prompt adherence.

At longer distances ADETAILER is recommended.

💕This is probably going to be the last version of the model for the moment as Im quite happy with the result. Thanks for downloading and I hope you enjoy it :)

☕If you want to support me feel free to do so on Ko-Fi :)

⚡️BUZZ FOR BEST IMAGES⚡️

## Trigger words

You should use `None` to trigger the image generation.


## Download model

Weights for this model are available in Safetensors format.

[Download](/LyliaEngine/ponyRealism_v21MainVAE/tree/main) them in the Files & versions tab.