Sign In

Photonic Fusion SDXL

331
3.5k
124
Updated: Apr 18, 2025
base modelgaytwinkmalemenrealistic
Verified:
SafeTensor
Type
Checkpoint Trained
Stats
1,907
533
1k
Reviews
Published
Feb 22, 2025
Base Model
SDXL 1.0
Hash
AutoV2
3D781844E3
A1111 Badge
StecFX's Avatar
StecFX

Welcome to Photonic Fusion SDXL Finalé!

I stopped posting my models to CivitAI but since this is my last checkpoint and my best work I decided to make an exception as I would like as many people as possible to experience it.

This is my grand finalé, I wanted my Photonic Fusion series to go out with a bang and I'm confident I succeeded. I have decided to hang up my hat for a while when it comes to checkpoints. I still have some LoRA's in the pipeline though. I hope you have enjoyed this AI journey with me and thank you for the love and support and just using my models all these years! I primarily made these for me but I am so happy so many of you have enjoyed them as well!!!

I'm super proud of this version.... it excels at NSFW especially male anatomy but now also female anatomy. It's a big improvement over my previous checkpoints in photo realism pushing SDXL to it's limits with male anatomy, realistic body hair/pubes, better teeth and tongues, the most realistic vagina's, realistic skin texture, better body variety and sizes for both sexes, improved prompt adherence and overall better little details that make the image magical. There is still a dose of my Pony model merged in so Pony Score tags work on this and are actually encouraged for best results. Obviously I advertise my Photonic Fusion series of checkpoints as male focused but this model now also excels at girls as well and obviously femboys too.

  • On A1111/ForgeUI I recommend DPM++ 3M SDE + Karras

  • Steps 30-40

  • CFG: 5-6.5

  • 768 x 1024 or 768 x 1100 Upscaled 1.5 - 2x

  • Hires Fix, 10-15 Steps, 0.30-0.40 Denoising, 1x-ITF-SkinDiffDetail-Lite-v1 is the best on A1111 and ForgeUI (also recommend 4x_NMKD-Superscale-SP_178000_G)

  • I suggest using ADetailer both for face face_yolov8n.pt and eyes mediapipe_face_mesh_eyes_only with appropriate prompting for full body images of people but it's not really necessary for close up shots.

If you find a better combination of settings again please share with the community in the comments below!

V1 is here on CivitAI and TensorArt, V2 is exclusively for MageSpace so feel free to try it out there, V3 is exclusive to TensorArt.

Also I don't believe in locking my models that are available on TensorArt or CivitAI behind a paywall like some do but I gladly accept donated --> Coffee and Beer <-- your donations help me to keep motivated to continue making new checkpoints and LoRA's for all our enjoyment! To those that are generous enough to throw me a few dollars I thank you! <3


UPDATE! V3 is out but is exclusively available on my TensorArt page HERE!

This is my first true SDXL successor to my SD1.5 Photonic Fusion male focused models. PLEASE READ BELOW!

Currently in beta testing! As you can see from the example images it can do some really amazing things, and create some incredible cocks both with foreskin and without, huge, average and even small. But it's also not consistent and rough around the edges when it comes to prompt adherence. This one has a bit of a learning curve to it, your favourite prompts when copy pasted into this checkpoint will likely give you vastly different results sometimes amazing and sometimes absolute crap lol, you will need to adjust your prompts to this specific model. There is a dose of my Pony model merged in so Pony Score tags work on this and are actually encouraged for best results as well as PAG. Obviously this is a male focused model but it's also great at girls with or without cocks and femboys as well.

I will continue to refine and improve the model over time. Please share your creations and please provide feedback both good and bad.

I'm still trying to figure out the best settings but these seem to work pretty good.

  • On A1111 I recommend DPM++ 3M SDE + Automatic

  • Steps 40-50

  • CFG: 3-6

  • Size: 512x768 or 768x512 or 1024x1024 or 768x1024

  • Hires: Denoising Strength: 0.30-0.50 + Hires upscale: x2, Hires steps: 13-20 and for Hires Upscaler I highly recommend: 1x-ITF-SkinDiffDetail-Lite-v1 or 8x_NMKD-Superscale_150000_G or R-ESRGAN 4x

  • I suggest using ADetailer both for face face_yolov8n.pt and eyes mediapipe_face_mesh_eyes_only

If you find a better combination of settings again please share with the community!

My original Merge pre-finetune includes a dose of both these amazing checkpoints...

bigASP https://civitai.com/models/502468?modelVersionId=991916

LUSTIFY! https://civitai.com/models/573152?modelVersionId=1569593

Also I gladly accept donated --> Coffee and Beer <-- your donations help me to keep making you checkpoints and loras, thank you!