This is a failed attempt at making a Top Down style/asset.
It makes some interesting pictures so I wanted to post it (also I didn't want to lose the file lol) but its a failure because it seems to not be able to make very many different things.
This is trained on almost entirely pictures from the game Sea of Star that DreamExplorer collected.
I've been thinking about why this isn't flexible for awhile but still haven't come up with any solutions.
Description
Trained on: 42 images
Training Model: AnyLoRA_noVae_fp16-pruned
flip_aug: True
Num of Repeats: 10
Unit is Epochs or Steps: Epochs
Number of Epochs or Steps: 10
Training Batch Size: 2
Network Dim: Dim=64; Alpha=32
Lora Creation Process took: 0:25:38.
Details
Downloads
1,391
Platform
CivitAI
Platform Status
Available
Created
12/27/2023
Updated
9/28/2025
Deleted
-
Trigger Words:
no humans, scenery, top down view
Files
Sea_of_Stars.safetensors
Mirrors
CivitAI (1 mirrors)