Nodes Browser

ComfyDeploy: How ComfyUI_HF_Inference works in ComfyUI?

What is ComfyUI_HF_Inference?

Unofficial support for Hugging Face's hosted inference.

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 ComfyUI_HF_Inference and select it
  5. Close the build step dialig and then click on the "Save" button to rebuild the machine

Hugging Face hosted inference nodes for ComfyUI

Unofficial ComfyUI nodes for Hugging Face's inference API

Visit the official docs for an overview of how the HF inference endpoints work

Find models by task on the official website

Installation

Clone and install dependencies

git clone https://github.com/bitaffinity/ComfyUI_HF_Inference custom_nodes/ComfyUI_HF_Inference
cd custom_nodes/ComfyUI_HF_Inference
pip install -r requirements.txt

Export HF_AUTH_TOKEN with one of your Hugging Face tokens

Run ComfyUI

HF_AUTH_TOKEN=hf_1111111111111111111111111111111111 python main.py

Nodes

[!WARNING] Inference API (serverless) requires a model 10GB or below and fails for random reasons on different models.

Text

Image