Klein 9B
Showcase images :
Distilled model
768x1152
Euler / normal / 4 steps / CFG 1
Lora strength of 0.8-1
Qwen:
v2: Retrained on a completely new and higher quality dataset (same one used for training the ZIT version)
ZIT:
Showcase images settings :
Lora strength 0.7 - 0.8
DDIM & Beta
10 steps & 1 CFG
1056x1584
'comic style image'
Long Markdown type prompt for some of them
Flux:
Trained on 30 images handpicked from the comic The adventures of Tintin
It's a style Lora, so it doesnt really know the characters (except maybe Tintin and his dog which Flux already knew a bit).
a bande dessinée style image of
with a weight around 0.8-1. (for Qwen it's 1)
Description
FAQ
Comments (5)
Big thanks for including the prompts. It is always cool when an author so loves their LORA that they post tons of example renders! Looks excellent.
It's frustrating when you open a Lora page and there is only one or two images
Nice results!
I'm using Flux1D-fp8 (only have 8GB VRAM) on ComfyUI and the results looks less good than yours.
Did your results are from fp16 version on Flux?
Many thanks!
I can't run Flux localy so I use TensorArt, and they say it's fp8.
Most of the time I generate images at 768x1152, then upscale it x1.5 (denoising strength around 0.5 and 4x_NMKD_Siax_200k as upscaler).
Thank you very much!