Type | |
Stats | 1,108 |
Reviews | (97) |
Published | Jun 6, 2023 |
Base Model | |
Training | Steps: 2,970 Epochs: 18 |
Trigger Words | skye \(valorant\) |
Hash | AutoV2 190ECD2A4C |
LoRA trained on Skye from Valorant
models used
anime: YabaL_Mix V3
2,5D: CamelliaMIx_2.5D - V2, DarkSun, HELLmix
realistic: Paragon V1.0
tags: skye \(valorant\), 1girl
optional tags: headband, fingerless gloves, body armor, single elbow glove, belt, pants, crop top
negative embeddings: bad-hands-5, ng_deepnegative_v1_75t, bad-artist, EasyNegative
I am now using Automatic1111 again, so you guys can follow the steps more easily and see the models I used without looking into the description.
The switch to the webui was harder than i thoght lol :D
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 : Hires. fix, Upscale by 2, 14 Hires steps, 4x-UltraSharp(for 2,5D & realistic), 4xAnimesharp(for anime)
CFG Scale : 5 to 9 (mostly 7)
LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.8)
Denoise strength upscaling: 0.4 to 0.7
for full body shots i recommend using: ComfyUI Impact Pack
or the a1111 alternative GitHub - Bing-su/adetailer