Model

Here you define the base model you use for training and the output model you want to achieve.

Base Model: either provide the the path of a saved base model or provide a link to it. Set the Model type (SD1.5 to SDXL1.0) as well. Model Output destination: set any local folder with its name and type (ex C:/OneTrainer/Output/xxx.safetensors) Data Type: 3 options are available at the moment, float16, bfloat16 and float32. They all have different precision and will consume more or less memory.

Note 1 : We need some more explaination on the data types. Noet 2 : We need a tab to say which dtype are relevant for each kind of training (embedding, fine tune, Lora, Fine tune VAE)