训练模型是anythingV5, 生成预览图用的模型是anyhentai20,背景不太好控制，建议多写一些背景提示词或提高其权重（素材没抠图的锅）人物特征提示词：1girl, solo, gloves, fingerless gloves, braid, black footwear, ahoge, navel, long hair, pants, black gloves, headset, goggles, goggles on head, midriff, brown hair, cloak, yellow eyes, crop top, standing, brown eyes, black pants, shoes, cape建议使用的反向提示词: holding, simple background, grey background, weapon,3d rendering,建议使用分层控制（LoRa Block weight)，预览图我使用的分层控制参数是AACC:1,1,1,1,1,0,1,1,1,1,1,0,1,0,0,0,0.5，

Files with SHA256: b818a4ef929a571b64b406e7e1849787ea3bdee20d7662513e424283e9bf9a0c