Pro Version of Realism Pony V4-V5-V6 now Available on My Patreon
Onsite generations are permanently available on these models:
š Realism_By_Stable_Yogi V3: https://civarchive.com/models/166609?modelVersionId=992946
To keep this model available for Image generation on Civitai, please place your bids here and Iāll make it available for onsite generation:
https://civarchive.com/auctions/featured-checkpoints
Join me on Patreon for exclusive perks and early access to unique resources.
To discuss custom LoRa's or models, feel free to connect on Discord.
š Like this model to keep me motivated and inspired to create more!
š¬ Drop a comment and let me know what you'd love to see next.
š Review this model to help me improve and make even better creations.
š Hit that notification bell to stay updated with my latest models and updates!
Important Usage Tips
Add Stable_Yogis_PDXL_Positives at the beginning of your prompt section.
Add Stable_Yogis_PDXL_Negatives-neg at the beginning of your negative prompt section.
Description
REALISM_PONY_V4_VAE_BY_STABLE_YOGI
Pro Version of Realism_v4_Fp16 and DMD2 is Available on My Patreon
Better with Realism, Skin Details, Lighting, dynamic poses, diverse faces. Please test and Share feedback.
Sampler Euler a, DPM2 a, DPM++ SDE, DPM++ SDE Karras, DPM++ 2M SDE-- SGM Uniform,
CFG Minimum 4+ maximum upto 7
Steps Minimum 18 , Recommended Steps 27
Resolutions: All SDXL Resolutions
Adetailer (Required)
High res Fix (Optional)
Denoise - 0.30
Minimum hires steps 5+
Minimum Upscale 1.5
Upscaler 4x-UltraSharp
Keep the VAE at Auto ( VAE Included )
Negative Prompts
(Download this negative embedding for best results) Stable_Yogis_PDXL_Negatives
Positive Prompts
(Download this positive embedding for best results) Stable_Yogis_PDXL_Positives
FAQ
Comments (32)
Sweet baby Jesus! The inpainting power on this thing is insane. Definitely an advance from V3
Really appreciate your feedback.
How do you do inpainting with non-inpaint checkpoints? It always fails for me
@GraphXĀ I use auto1111 and use these basic settings every time: Only masked; Only masked padding pixels 12; sampling steps 45-49; Batch count 3-5; Denoising strength 3.2-3.7. Those settings always work for me. I hope it helps.
This checkpoint can handle in-paint.
@_Soda_ my denoising strength slider cant go to 3 it stops at 1 for me
with the latesy pony version, I often get bad nipples when i2i, no idea why
@Stable_YogiĀ And here I was using the old inpaint versions like a chump
The images this model creates are outstanding, but i cant help to notice that any time i want the head to be facing a diferent direction is kinda dificult to have it that way, it always has this slight tilt to some side, have any of you notice this before and is there any way to solve this?
I use v1,v2, v3 regularly, they are good for different things.
The biggest improvement with v4 is skin texture, there's definitely a lot more realism with v4!
Thanks for the feedback !
Any possibility of a version of this realism model for the NoobAI base model? š
Sorry, Nothing planned for NoobAi yet.
i liked your v4_Fp16, why can't we use it in the generator anymore?
Which version ?
Currently V3 is setup for Image generations.
The v4_Fp16 version, before the v3 was the only one version to generate images in civitai.
@mrfeetĀ will make it available in next weeks bidding.
@Stable_YogiĀ Thank you!
If thereās another model youād like to generate images with, just place a bid here and Iāll make it available for onsite generation:
https://civitai.com/auctions/featured-checkpoints
@Stable_Yogi the v4_Fp16 version is perfect, ty
If thereās another model/version youād like to generate images with, just place a bid here and Iāll make it available for onsite generation:
https://civitai.com/auctions/featured-checkpoints
What is dmd2 and fp16? what is the difference?
I think fp16 is floating pointer accuracy, where dmd2 is double floating pointer accuracy... even though I have a hart time noticing any differences...
dmd2 is a Lora, merged for faster image generation. Like 4 steps.
Its a Lightening Version.
It produces tremendous high contrast NSFW images (using Fooocus). However, it doesn't seem to like Outpainting (even on images it has created itself).
The additions it outpaints are all pixelly and grainy.
Is there a Fooocus setting in debug I need to change to stop this happening.
I am not sure about the settings. I don't use focus much. This works perfect with A1111 and Comfy. Every UI has its own perks and issues.
do you mean standard or DMD2 version ?
@mariedoll123Ā The filename is realismByStableYogi_v5XLFP16
@JohnnyWu22Ā try changing the sampler to Eular ancestral / normal , can give a softer result
Beyond awesome
Absolutely love this lora
Thanks for the feedback.
Details
Files
realismByStableYogi_v40FP16.safetensors
Mirrors
oppathebear_v10.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
realismByStableYogi_v40FP16.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.