Test LoRa trained on 32 images in OneTrainer Trigger - "The low angle image" Works best with womenModel used to generate images V1 version is AcornIsSpinning FP8&amp;NF4 (Full steps) Model used to generate images V2 version is PixelWave dev03

Low angle photography

Trained on 1024px - better skin, updated captions If you see horizontal/vertical lines - you can disable layers in LoRa (<a target="_blank" rel="ugc" href="https://www.reddit.com/r/StableDiffusion/comments/1g7b76e/a_possible_fix_for_the_gridlines_effect_when/">https://www.reddit.com/r/StableDiffusion/comments/1g7b76e/a_possible_fix_for_the_gridlines_effect_when/</a>) or upscale without lora by &lt;0.40

Files with SHA256: 2f32bb4b46037d388fe8e1f13e1d8eef9c8b9f969d63276c450cceb023f4b7c8