Nodes Browser

ComfyDeploy: How Recognize Anything Model (RAM) for ComfyUI works in ComfyUI?

What is Recognize Anything Model (RAM) for ComfyUI?

This is an image recognition node for ComfyUI based on the RAM++ model from [a/xinyu1205](https://huggingface.co/xinyu1205). This node outputs a string of tags with all the recognized objects and elements in the image in English or Chinese language. For image tagging and captioning.

How to install it in ComfyDeploy?

Head over to the machine page

  1. Click on the "Create a new machine" button
  2. Select the Edit build steps
  3. Add a new step -> Custom Node
  4. Search for Recognize Anything Model (RAM) for ComfyUI and select it
  5. Close the build step dialig and then click on the "Save" button to rebuild the machine

Recognize Anything Model (RAM++) for ComfyUI

Image Tagging and Captioning

The counterpart to Segment Anything Model (SAM)

This is an image recognition node for ComfyUI based on the RAM++ model from xinyu1205.

Installation

Git clone this repository inside the custom_nodes folder or use ComfyUI-Manager and search for "RAM".

ℹ️ In order to make this node work, the "ram" package need to be installed. To do so, open a terminal inside the python_embedded folder within your ComfyUI installation:

python.exe -m pip install git+https://github.com/Hangover3832/recognize-anything.git

Furthermore you need to download the RAM, RAM++ and tag2text models and place it in the /ComfyUI/models/rams/ folder or use the ComfyUI-Manager model downloader. You can also configure the location in 'extra_model_paths.yaml' in the Confyui folder.

Updates

  • Added RAM and tag2text models.
  • Support for the 'extra_model_paths.yaml' configuration.

To Do