Community DiT
This is a model with a Style LoRA merged into the Anima Base or AnimaYumeIt may not be suitable as a base model for training. It is recommended to use Anima Base or AnimaYume as the base for training.※※※ This checkpoint is not an SDXL-based model. ※※※Generation settingsresolution1024x1024, 896x1152, 1152x896, etc.1535x1536, 1344x1728, 1728x1344, etc.steps30～50(8～12 v2.0 Turbo)CFG3.0 ～ 5.0(1～2 v2.0 Turbo)PromptQuality tagmasterpiece, best quality, good quality, score_9, score_8, score_7, highly detailed, highres, absurdres, newestSince this model is merged with a style LoRA, the use of @artist —which significantly affect the visual style—is not recommended.Negative tagworst quality, low quality, score_3, score_2, score_1, old, oldestIncluding the number of fingers and toes in the prompt can help improve the results.Positive5-fingers, 1-thumb, 5-toesNegative6-fingers, 6-toesPlease refer to this as well.Version 1.2 has been discontinued.