<h1 id="lora-details-m7is2hmev">LoRA Details</h1><ul><li>Test tune of ~17K images trained at 1024x1024 with buckets using captions from CogFlorence-2.2 at batch size 4 for 4 epochs.</li><li>All images were manually filtered to the best of my ability to avoid low quality/watermarks/AI/etc.</li><li>All images are originally at least 1024x1024 total pixels.</li></ul><h1 id="reason-for-creating-this-lora-8ot71luqd">Reason for creating this LoRA</h1>I wanted to throw some fairly high quality but "general use-case" images at a model to use as a base so it can be further trained on more specific tasks in the future.

SDXL - XZHQ

This version does not contain much nudity, and may be biased towards women. I am working on increasing the dataset size and variety, and optimizing the training settings if needed. There is a v1, but it was trained at batch size 1 for 1 epoch with the text encoder LR at half the amount v1.1 was trained at, so I don't feel it is worth publishing.

Files with SHA256: ef574d7340924758f31c5a4d9f8690b31c1380f888b94b643c43dd5bafb2adde