<h3 id="flux:-ysae6q2gt"><strong>Flux</strong>:</h3><p>Trained on FLUX1.dev with 237 images</p><p>Tested with ComfyUI and Forge...</p><p></p><h3 id="sdxl:-0wabfksfw"><strong>SDXL</strong>:</h3><p>Trained on SDXL 1.0 vae fix with 281 images.</p><p>The result of another test training trying to reduce the file size without compromising (too much) on likeness/quality and without resizing the model afterwards...</p><p>And again, there is still a lot of room for improvement..</p><p>No VAE, hiRes fix or refiners were used for the example images...<br /></p>

Files with SHA256: be387db2a651cfaf1df50423a993501499d6439e4c36ff1b093b59fb283c1d41