LoRA trained on Blitzcrank from League of Legends
models used
anime: YabaL_Mix V3(image 10), MeinaAlter, PrimeMix - anime(image 2)
2,5D: HELLmix(image 3,5), DarkSun(image 6), MeinaUnreal(image 4,7,8), ReV Animated(image 1)
realistic: fantasticmix(image 9,11,12)
tags: blitzcrank
optional tags: robot, mecha armor, mecha, spikes
negative embeddings: embedding:bad-hands-5, embedding:ng_deepnegative_v1_75t, embedding:bad-artist, embedding:EasyNegative
Recommended settings:
Make sure you are aware on the usage instructions of LORA
VAE: vae-ft-mse-840000-ema-pruned (for realistic models & 2,5D), kl-f8-anime2, blessed2.vae (for anime models)
Sampler: DPM++ SDE Karras or DPM++ 2M Karras
Steps: 20 to 35
Clipskip: 2
Upscaler : img2img in comfyui using 4x-UltraMix_Smooth
CFG Scale : 5 to 9 (mostly 7 and 8)
LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.8)
Denoise strength for img2img: 0.4 to 0.7
for the pictures i used comfyui
Description
FAQ
Comments (7)
That pretty cool đź‘Ť
May I ask what is your method of making loras please? Any guide or video that I could follow?
It depends on the GPU you have. If your GPU has at least 6GB of VRAM, you can use the bmaltais/kohya_ss web UI, which I also use. However, if you have a GPU with less VRAM or just a slow GPU, I would suggest using Google Colab, where you can borrow some Google GPUs.
This guide is pretty good: THE OTHER LoRA TRAINING RENTRY
And for Google Colab training, I suggest this: [Guide] Make your own Loras, easy and free - colabs
@AhriMain much obliged it!
Which Model do you use to train it ?
can you do tahm kench lora pleasE?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.