
This LoRA was trained on MidJourney images of Painted Vinyl Statues combined with my Personalized Style. Initially, I planned this as a "mini" concept for Flux, consisting of only 60-80 images, but after building the diverse dataset, I expanded it to include 120 DnD miniatures.
Despite the original training data being specific, the model turned out to be much more flexible than expected, which is why I chose not to limit its name to "DnD Vinyl Statue." However, it still shows a strong preference for characters and monsters.
Usage:
-
Flux Trigger: ArsFigurine, Vinyl statue.
-
Strength Range: Works well between 0.2 - 1.5, depending on the LoRA stack used.
-
For best results, I recommend starting between 0.8 - 1.
-
Above 1.5 the quality starts to degrade really fast, but you can get interesting results up to 2.
描述:
All images in the Showcase Gallery were created using this workflow and the Flux Q4 Dev model on a machine with 8GB VRAM.
https://civitai.com/models/658639/super-simple-gguf-quantized-flux-lora-workflow
训练词语: ArsFigurine, Vinyl statue
名称: 781837_training_data.zip
大小 (KB): 199783
类型: Training Data
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: VinylStatueFlux.safetensors
大小 (KB): 37506
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success