A redux from the old dataset of Indonesiadoll likeness Lora
Fonte do modelo
Descrição da fonte
A redux from the old dataset of Indonesiadoll likeness Lora
I tried making the kohya ss lycoris training to work, and quite happy with the result.
Recommended weight starting from as low as 0.3 upto 1.0. Use your preferred sampler (2M, SDE or 2M SDE Karras).
i also recommend to Use hires.fix.
i find it not suitable with asian checkpoint merges. (As usual i train this lycoris with SD1.5).
I'd be happy to see your creation! keep it SFW though
Fontes
1 fonteArtefatos de modelo
1 artefatoindonesiadoll-lyco-ver1.1.safetensors
safetensors · 74,4 MB · SHA-256 1c545799ed23…4d05 · Civitai
BaixarTrechos de fonte
1 trechoSource context: 1036 downloads · Type LoCon · Base model SD 1.5