🚀 用於Flux的兇狠高大龍LoRA模型
本項目是一個基於LoRA技術的模型,專為Flux.1-dev模型微調,能夠生成更加兇狠、高大的龍形象,解決了原模型在表現龍的規模和姿態方面的難題。
🚀 快速開始
激活方式
無需觸發詞,可使用以下提示詞:
- 一隻巨大凶狠的紅龍高聳在一個小騎士之上
- 一座山峰上的巨型黑龍
推薦訓練輪數
使用12 - 15輪訓練的文件(已提供),建議從14或15輪開始。
LoRA強度
Clip和模型強度設置在0.8到1.0之間。
分辨率
建議使用垂直分辨率(例如768x1344)以展現規模感。
✨ 主要特性
本LoRA模型針對Flux.1-dev進行微調,能夠生成更兇狠、更具威脅性且體型更大的龍,使其高聳於人物或風景之上,有效解決了Flux.1-dev在表現規模和姿態方面的挑戰。
📦 安裝指南
文檔未提及安裝步驟,暫不提供。
💻 使用示例
基礎用法
使用如下提示詞即可生成相應的龍圖像:
massive menacing red dragon towering over a tiny knight
高級用法
可根據不同的場景和需求,調整提示詞的細節,如龍的顏色、姿態,人物的穿著等,以生成更符合預期的圖像。例如:
A photo-realistic shoot from a profile camera angle about a young girl with long, flowing blonde hair standing in front of a large, menacing dragon. the girl, who appears to be in her early twenties, is facing away from the camera, with her back to the viewer. she is wearing a red dress with ruffled sleeves, and her hair is styled in a braid. the dragon, with its sharp teeth and orange scales, has a menacing expression, and its eyes are focused intently on the girl. the background is blurred, with rocks and greenery visible, and the lighting is dramatic, casting shadows on the dragon's face and body. the overall mood is dark and moody, with a focus on the character and the dragon.
📚 詳細文檔
模型描述
本LoRA模型針對Flux.1-dev進行微調,旨在生成更兇狠、更具威脅性且體型更大的龍,使其高聳於人物或風景之上,解決了Flux.1-dev在表現規模和姿態方面的挑戰。
標題風格
示例標題風格展示瞭如何通過詳細的描述,營造出特定的場景氛圍,如:
A photo-realistic shoot from a profile camera angle about a young girl with long, flowing blonde hair standing in front of a large, menacing dragon. the girl, who appears to be in her early twenties, is facing away from the camera, with her back to the viewer. she is wearing a red dress with ruffled sleeves, and her hair is styled in a braid. the dragon, with its sharp teeth and orange scales, has a menacing expression, and its eyes are focused intently on the girl. the background is blurred, with rocks and greenery visible, and the lighting is dramatic, casting shadows on the dragon's face and body. the overall mood is dark and moody, with a focus on the character and the dragon.
關鍵訓練參數
參數 |
詳情 |
引擎 |
kohya |
訓練輪數 |
17 |
重複次數 |
6 |
步數 |
1887 |
分辨率 |
1024 |
Unet學習率 |
0.00020 |
學習率調度器 |
cosine_with_restarts |
優化器 |
AdamW8bit |
網絡維度(秩) |
64 |
網絡Alpha |
32 |
🔧 技術細節
文檔未提供具體的技術實現細節,暫不展示。
📄 許可證
本模型以MIT許可證發佈,可自由使用、修改和分發,但請註明來源。
模型下載
本模型的權重以Safetensors格式提供。
點擊下載(在文件與版本標籤中)。