Flux Claude Art

🚀 Flux-Claude-Art

Flux-Claude-Art 是一個文本到圖像的模型，可根據輸入的文本描述生成相應的卡通或數字繪畫風格圖像。它基於特定的基礎模型進行訓練，通過特定的觸發詞來生成圖像，為用戶提供了便捷的圖像生成體驗。

🚀 快速開始

環境設置

import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "strangerzonehf/Flux-Claude-Art"
trigger_word = "claude art"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

觸發詞使用

你應該使用 claude art 來觸發圖像生成。

模型下載

此模型的權重以 Safetensors 格式提供。 點擊下載，可在 “Files & versions” 標籤中進行操作。

✨ 主要特性

• 文本到圖像生成：根據輸入的文本描述生成對應的卡通或數字繪畫風格圖像。
• 特定觸發詞：使用 claude art 觸發圖像生成。

📦 安裝指南

安裝步驟已在 “快速開始” 部分的環境設置中給出，請參考相應的代碼示例進行安裝。

💻 使用示例

基礎用法

import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "strangerzonehf/Flux-Claude-Art"
trigger_word = "claude art"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

# 示例文本輸入
text_input = 'Claude Art, An animated cartoon image of a man in a white collared button-down shirt and a brown apron. The mans arms are crossed in front of him, and he is holding a tablet in his right hand. He has a serious expression on his face, and his eyes are squinted. His hair is dark brown, and the background is a light green, with the letters U visible on the left side of the image.'
image = pipe(text_input).images[0]
image.save("generated_image.png")

📚 詳細文檔

圖像示例

圖像參數

圖像處理參數

最佳尺寸與推理

📄 許可證

本模型使用 creativeml-openrail-m 許可證。