Minecraft-Styled Loras for XL: Styling & Comparison
Based on pre-rendered Minecraft AI-generated photos, utilizing our custom SDXL models.
Lora Civitai
Lora Lokon (tba)
Lora FA (tba)
Lora Standard (tba)
Description
LyCORIS - Text Encoder only
Resolution: 1024x1024 Architecture: stable-diffusion-xl-v1-base/lora
Network Dim/Rank: 32.0 Alpha: 32.0
Module: lycoris.kohya : {'conv_dim': '32', 'conv_alpha': '32', 'algo': 'lora'}
Text Encoder (1) weight average magnitude: 1.763796416299954
Text Encoder (1) weight average strength: 0.00881813765564231
Text Encoder (2) weight average magnitude: 1.827049986849515
Text Encoder (2) weight average strength: 0.007027856117622828
No UNet found in this LoRA
FAQ
Comments (2)
So these are different style loras with varying intensities of the minecraft style? Is that it?
Indeed, this initiative began as an experiment to juxtapose the CivitAI trainer with other available options. The first one was trained on the CivitAI official trainer, the second one on normal LORA, the third one on LORA FA with only a text encoder, and the fourth one on Lycoris. I encourage you to try them all out and see which one works best for you. You can also combine them to create your own unique style. A heavier trained version is coming soon, and I would appreciate it if you could leave a note about what works for you and which LORA you would like to see further trained. Posting a few mash up examples with other loras and random prompts for the test. so far it does blend well