V4.5回覆動慢風格更鮮豔的顏色和細緻的臉孔Slow and repetitive styleBrighter colours and more detailed faces.V3_photoartThe actual effect is closer to 2.5D-3D game cg and its mergerd with SDXL photo-realistic relationship and the merging of the art department's big models, which reduces the recognition of the anime characters and brings better support for the real portraits and art-like LORA.實際效果更接近2.5D-3D game cg和SDXL 真實照片係還有藝術系的大模型合併,降低了對動漫人物的辨識度帶來對真實人像 藝術類LORA更好的支援Sugest : WebUi re-ForgeSteps: 12Sampler: LCM,Schedule type: Automatic,CFG scale: 1.5-2.5, (4 without FreeU, 1.5-2.5 with FreeU)Model: UnstableInkDreamXL-V3_photoart,Clip skip: 2,Lora : negXL:0.5V3.3Stable output with art LORA.Slightly eliminates the influence of anime and mangaSugest : WebUi re-Forge Steps: 12Sampler: LCM,Schedule type: Automatic,CFG scale: 1.5-2.5, (2 without FreeU, 1.25-1.5 with FreeU)Model: UnstableInkDreamXL-V,Clip skip: 2,Lora : negXL:0.5V2.0Minor ImprovementsAdd some artistic LORA adjustmentsSugest : WebUi re-Forge Steps: 8-16, (10-16 without FreeU, 6-8 with FreeU)Sampler: LCM,Schedule type: Automatic,CFG scale: 1.5-2.5, (2 without FreeU, 1.25-1.5 with FreeU)Model: UnstableInkDreamXL-V02,Clip skip: 2,Lora : negXL:0.5V1.0My first test of illustration checkpoint mergeOld sd1.5 checkpoint still well for style paintings,But ....https://civitai.com/models/1540Using WEBUI forge,with LCM sampling method and CFG 3~3.5,step 8-20examplesClip skip 2 LCM/Autometic 8-20 stepsCFG 3-3.5640*896 resolutions or higherpromptmasterpiece, best_quality,absurdres, highres,negativeworst_quality, bad_quality, bad_hands,very_displeasing,lowres,extra_digit,fewer_ digits,jpeg ,watermark, instagram_username, twitter_username, patreon_username,frame,comic_fram, artifacts, signature, username, reference,