Based on Anima Preview 2<a target="_blank" rel="ugc" href="https://huggingface.co/circlestone-labs/Anima">Anima</a> is a 2-billion parameter text-to-image model developed by <a target="_blank" rel="ugc" href="https://huggingface.co/circlestone-labs">CircleStone Labs</a> in collaboration with Comfy Org. Built on the <a target="_blank" rel="ugc" href="https://research.nvidia.com/labs/dir/cosmos-predict2/">NVIDIA Cosmos</a> architecture and utilizing the Qwen text encoder, it serves as a specialized foundation for high-end anime and artistic generation.<h3 id="model-recipe">Model recipe</h3>Merge of <a target="_blank" rel="ugc" href="https://civarchive.com/models/2385278/animayume?modelVersionId=2798563">AnimaYume_v0.3</a> + <a target="_blank" rel="ugc" href="https://civarchive.com/models/2383017/anima-cat-tower?modelVersionId=2790168">AnimaCatTower_v0.4</a> + own trained LoRAsMerging the models:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/fba8cf05-b711-4aa5-9f18-f840ae7a7ac8/original=true/fba8cf05-b711-4aa5-9f18-f840ae7a7ac8.jpeg" />Merging the loras:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/76ffd0fb-cc21-4606-9286-6e5c1a17cdf8/original=true/76ffd0fb-cc21-4606-9286-6e5c1a17cdf8.jpeg" />Workflows are embedded in the images.<h3 id="personal-recommended-settings">Personal Recommended Settings</h3><ul><li>Steps: 25</li><li>CFG Scale: 5.0</li><li>Sampler: <code>euler_ancestral</code></li><li>Scheduler: <code>simple</code></li></ul><a target="_blank" rel="ugc" href="https://www.comfy.org/workflows/image_anima_preview-cc053bb55df6/">ComfyUI Workflow</a>

Files with SHA256: d18dec3149bd02c02090da2611b5d6a9ccb10d0c887a286df9a76266b2e67fd6