This series of the models was started (v0.4) from DucHaiten-Pony-Real (beta) by DucHaiten.
Version v1.0 was trained using a LoRa mode in One trainer. Next version are trained in Fine Tuning mode (full SDXL training).
Up to v1.0: Photorealism, lighting and anatomy related LoRas from CinEro series are integrated into this model in order to amplify cinematic render.
Goal of training was to improve to stability of anatomy and make the picture more realistic and moody / cinematic.
Version 1.1: shifted to fantasy / sci-fi themes with less realistic and more contrast / unstable rendering.
Version 1.2 [SOON]: is v1.1 further development with shift to horror / tentacles theme.
PS: be responsible and don't abuse this unrestricted model.
Description
CAUTION: this version is more sensitive on CFG. CFG less than 5.5 is optimal.
Added 3 epochs of training on 3 datasets: mountainous landscapes, moody sky and sfw model photosession on nature.
Improvements: more rich backgrounds with more smaller objects, more vibrant colors and slightly higher contrast, added subtle analog film grain effect.
FAQ
Comments (8)
Version CinEro_Pony_v0.9a.fp16 is good at detailing, but it is too sensitive on CFG. It sometimes gives unstable, overcooked results. I'm suggesting to fix the contrast / instability issue. But I don't want it to become too smooth.
I feel the same :( Thx 4 making me noticed!
I've uploaded v0.9d.
CFG sensitivity and contrast slightly reduced.
Version v0.9d is much better by my preferences... CFG issue fixed. Now it is much better, but now bokeh is strong.
Can be fied by adding "bokeh and shallow dof" to negatives.
Does the model generate Asian faces?
If you prompt it explicitly, then yes
Hello! I made a comparison of Pony models. 1 test prompt so far. More ongoing.
https://civitai.com/articles/6491
Comments / suggestions are welcome.
Thanks a lot for comparison. Will read carefully later.. So far seem quite objective.