Version for Anima Base 1.0 -&gt; ⚠️CHECK VERSION 1.0 UPLOADED IMAGES FOR TRAINING DATA PROMPTS ⚠️<hr />For Krea2 :⚠️CHECK ANIMA VERSION 1.0 UPLOADED IMAGES FOR TRAINING DATA PROMPTS ⚠️Use 0.9 or 0.8 strenght for the model to be more flexible.For the display generations Textfusion lora was used with 1.0 weight.<a target="_blank" rel="ugc" href="https://huggingface.co/Quiho/Krea2_TextFusion_Refusal-Reduction_LoRA_v1.0_lora/tree/main">https://huggingface.co/Quiho/Krea2_TextFusion_Refusal-Reduction_LoRA_v1.0_lora/tree/main</a><hr />2500 steps version of the lora, if there is interest i can release more flexible versions with less steps. The model is perfectly fine but you can always use 0.9 or 0.8 strenght for the model to be more flexible.ALL THE GENERATIONS I POSTED FOR VERSION 1.0 ARE RECREATIONS OF THE TRAINING DATA IMAGES WITH THE SAME TRAINING DATA PROMPTSTriggerword: inusen_styleAlmost all of them: light brown skin -&gt; You can use pale skin, dark skin for other styles<hr /><h3 id="characters">👤 Characters</h3>linda_inu, brown hair, long hair, eyes with black pupilsgloria_inu, long hair, blonde hair, teal color irisjoey_inu, 1boy, short hair, light brown hairshort hair, dark brown-reddish hair, red wavy hair tied in a bunlong hair, orange hair, green eyes. A young woman with long, wavy, orange hair adorned with a large purple bowblack hair, half-moon spectacles, blue eyes with black dot pupils. A woman with black shoulder-length hairblack hair, long hair, dark skin, eyes with black pupils. A voluptuous, dark-skinned woman<h3 id="used-generation-settings">⚙️ Used Generation Settings</h3>For examples i used the following settings based on the training parameters:<ul><li>Sampler: <code>euler_ancestral</code></li><li>Steps: <code>30</code></li><li>CFG Scale: <code>4.0</code></li><li>Lora weight: <code>1.0</code></li><li>Global Trigger Word: You must include inusen_style at the beginning of your prompts.</li></ul>

Inusen Style (Milftoon Drama)

Version for Krea2 -&gt; ⚠️CHECK ANIMA VERSION 1.0 UPLOADED IMAGES FOR TRAINING DATA PROMPTS ⚠️Use 0.9 or 0.8 strenght for the model to be more flexible.For the display generations Textfusion lora was used with 1.0 weight.<a target="_blank" rel="ugc" href="https://huggingface.co/Quiho/Krea2_TextFusion_Refusal-Reduction_LoRA_v1.0_lora/tree/main">https://huggingface.co/Quiho/Krea2_TextFusion_Refusal-Reduction_LoRA_v1.0_lora/tree/main</a>ALL THE GENERATIONS I POSTED FOR ANIMA VERSION 1.0 ARE RECREATIONS OF THE TRAINING DATA IMAGES WITH THE SAME TRAINING DATA PROMPTSTriggerword: inusen_style<s>Almost all of them: light brown skin -&gt; You can use pale skin, dark skin for other styles</s>Not needed for Krea2👤 Characterslinda_inu, brown hair, long hair, eyes with black pupilsgloria_inu, long hair, blonde hair, teal color irisjoey_inu, 1boy, short hair, light brown hairshort hair, dark brown-reddish hair, red wavy hair tied in a bunlong hair, orange hair, green eyes. A young woman with long, wavy, orange hair adorned with a large purple bowblack hair, half-moon spectacles, blue eyes with black dot pupils. A woman with black shoulder-length hairblack hair, long hair, dark skin, eyes with black pupils. A voluptuous, dark-skinned woman⚙️ Used Generation SettingsFor examples i used the following settings based on the training parameters:<ul><li>Sampler: <code>er_sde</code></li><li>Steps: <code>8</code></li><li>CFG Scale: <code>1.0</code></li><li>Lora weight: <code>0.9-1.0</code></li><li>Global Trigger Word: inusen_style (probably not needed)</li></ul>

Files with SHA256: ebf57e27d12c8d5854302524cd94c09e1a0766c0e4c5492bedf4cdb922aa5bbb