Read the "About this version" section to know what changes I made to the model.TL;DR: This model is a direct fine-tuning from AnyLoRA, unlike AingDiffusion, which merges multiple models and some personal datasets.The dataset used to fine-tune the model is still small, but already significantly affects the model.The model is still in development, and I will significantly update the dataset.===ElixirProject is a direct fine-tuning from AnyLoRa. This model is capable of generating high-quality anime images.<h3 id="heading-49">Guide to generating good images with this model</h3><ul><li>Use any negative embeddings (I recommend my stuff, <a target="_blank" rel="ugc" href="https://civarchive.com/models/163204/aissist-negative-embedding">AissistV2</a>, which I develop for my models.)</li><li>The recommended sampler are "DPM++ 2M Karras" and "DPM++ 2M SDE Karras"</li><li>Hi-res fix is recommended. The recommended upscaler is "4x-AnimeSharp"</li><li>Set clip skip to 2 (recommended, fine-tune dataset is trained on clip skip 2), ENSD (eta noise seed delta) to 31337 [to replicate image], and eta (noise multiplier) for ancestral samplers to 0.667.</li></ul>

ElixirProject

I've retrained the dataset using the Prodigy scheduler, and the result seems better than v1.3. I'm also adding a new dataset to the model. So there are 2 datasets right now.

elixirproject_v14.safetensors

v1.4

v1.6

v1.5

v1.3

v1.1

v1.0

E****rProject

I've retrained the datase* ***ng the Prodigy scheduler, and the result seems better than v1.3. I'm also adding a new dataset to the model. So there are 2 datasets right now.

Files with SHA256: 2a32fe5be7127e02becc1cb89e41df51ae737c438ea4a1105b9a2de0bf6f197f