comfyui-claude

comfyui-claude
★ 6

图像描述文本转换Anthropic ClaudeComfyUI 插件
将Anthropic的Claude模型集成到ComfyUI中，用于自动生成图像描述和对文本进行风格或语义转换，便于在可视化工作流中加入强大的NLP能力。
💡 在ComfyUI中用Claude为图片生成描述或对文本进行风格/语义转换。
🍴 5 Forks💻 Python🔄 2025-04-10
📦
网盘下载
复制链接后前往夸克网盘下载
https://pan.quark.cn/s/af9fbf81e746
📦 requirements.txt
anthropic>=0.36.0
pillow>=10.2.0
comfyui_types>=1.3.0
📄 README

ComfyUI and Claude

A set of custom nodes that are using Anthropic’s Claude models for describing images and transforming texts.

Setup

You can find the node in the ComfyUI package registry via its name “ComfyUI

Claude” and install it from there. Alternatively, clone the repository into your

custom nodes folder and install the requirements:

git clone https://github.com/tkreuziger/comfyui-claude.git ./ComfyUI/custom_nodes/comfyui-claude
python3 -m pip install -r ./ComfyUI/custom_nodes/comfyui-claude/requirements.txt

Then restart ComfyUI.

Requirements

You need an Anthropic API key that you must fill in the nodes. Learn more about

this here.

Included nodes

  • DescribeImage: Takes an image as input and returns a textual description of

    • it.

  • CombineTexts: Combine two texts into something new with the help of Claude.
  • TransformText: Transforms an input text into some other text, ideal for

    • rephrasing prompts or similar.