ComfyUI Gemini API Node

A custom node for ComfyUI that integrates Google's Gemini AI models for image analysis and description. This node allows you to send images to Gemini API and get AI-generated descriptions or analysis based on your prompts.

Updated to latest version. Added support latest version of gemini models.

Features

• Support for multiple Gemini models:
  • gemini-2.5-pro-exp-03-25
  • gemini-2.0-flash-exp
  • gemini-2.0-flash-thinking-exp-01-21
• Configurable safety filters
• Streaming response support
• Custom API key configuration
• Image to text generation

Rate limits

1)gemini-2.5-pro-exp-03-25:

-Requests per minute (RPM) - 5

-Requests per day (RPD) - 1,000,000

-Tokens per minute (TPM) - 25

2)gemini-2.0-flash-exp

-Requests per minute (RPM) - 10

-Requests per day (RPD) - 1,000,000

-Tokens per minute (TPM) - 1500

3)gemini-2.0-flash-thinking-exp-01-21

-Requests per minute (RPM) - 10

-Requests per day (RPD) - 1,000,000

-Tokens per minute (TPM) - 1500

Installation

1. Navigate to your ComfyUI custom nodes directory:

cd ComfyUI/custom_nodes/

2. Clone this repository:

git clone https://github.com/Merserk/ComfyUI-GeminiAPI-Latest.git

3. Install the required dependencies:

pip install -r requirements.txt

4. Create a config.json file in the node directory with your Gemini API key:

{
"GEMINI_API_KEY": "your-api-key-here"
}

Usage

1. Start ComfyUI
2. Find the "Gemini API" node under the "Gemini" category
3. Connect an image input to the node
4. Configure the following parameters:
   • prompt: Your text prompt for image analysis
   • model_name: Select the Gemini model to use
   • api_key: (Optional) Override the default API key
   • stream: Enable/disable response streaming
   • safety_filter: Enable/disable content safety filters

Configuration

You can configure the API key in two ways:

1. Add it to config.json file
2. Input it directly in the node parameters