Not an easy LoRa to work with, but it can work.
v2 is a slight improvement, but I still think I can do better. I think the training data needs higher quality.
Description
PICS: 131
BLIP: 10 - 75
network_dim: 64
network_alpha: 32
noise_offset: 0.05
steps: 10% 220/2200 [05:17<47:33, 1.44s/it, loss=0.0935]
steps: 20% 440/2200 [10:42<42:50, 1.46s/it, loss=0.0889]
steps: 30% 660/2200 [16:07<37:37, 1.47s/it, loss=0.0894]
steps: 40% 880/2200 [21:32<32:18, 1.47s/it, loss=0.0898]
steps: 50% 1100/2200 [26:56<26:56, 1.47s/it, loss=0.0894]
steps: 60% 1320/2200 [32:21<21:34, 1.47s/it, loss=0.0905]
FAQ
Details
Downloads
275
Platform
CivitAI
Platform Status
Deleted
Created
5/1/2023
Updated
5/11/2026
Deleted
5/23/2025
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.