- O/OpenAI
能力有限，目前仅支持在config.json中配置gpt模型
有其他问题请在issue中报告
Thank you to all the valuable contributors. Kindly submit any pull requests to the development branch instead of the main branch. Your efforts are greatly appreciated.
Extra nodes to be used in ComfyUI, including a new ChatGPT node for generating natural language responses.
ComfyUI is an advanced node-based UI that utilizes Stable Diffusion, allowing you to create customized workflows such as image post-processing or conversions.
Download the zip file. Extract to ..\ComfyUI\custom_nodes. Restart ComfyUI if it was running (reloading the web is not enough). You will find my nodes under the new group O/....
- quality of life will auto update each time you run comfyUI
- when you run comfyUI, the suit will generate a config file The file looks like this :
{ "autoUpdate": true, "branch": "main", "openAI_API_Key": "sk-#################################" }
- if you want to stop autoUpdate edit
config.jsonset "autoUpdate": false
This node harnesses the power of chatGPT, an advanced language model that can generate detailed image descriptions from a small input.
- you need to have OpenAI API key , which you can find at https://beta.openai.com/docs/developer-apis/overview
- Once you have your API key, add it to the
config.jsonfile - I have made it a separate file, so that the API key doesn't get embedded in the generated images.
- load_openAI:load openAI module
- Chat_Message: creates a message to be sent to chatGPT
- combine_chat_messages : combine 2 messages together
- Chat completion: send the messages to ChatGPT and receive answer
- create image
- variation_image
This set of nodes adds support for string manipulation and includes a tool to generate an image from text.
- Concat String: This node combines two strings together.
- Trim String: This node removes any extra spaces at the start or end of a string.
- Replace String : This nodes replace part of the text with another part.
- Debug String: This node writes the string to the console.
- Debug String route: This node writes the string to the console but will output the same string so that you can add it in middle of a route.
This node generates an image based on text, which can be used with ControlNet to add text to the image. The tool supports various fonts; you can add the font you want in the fonts folder. If you load the example image in ComfyUI, the workflow that generated it will be loaded.
- saveTextToFile: this node will save input text to a file "the file will be generated inside /output folder"
"node soup" which is a collection of different values categorized under different terminologies that you can use to generate new prompts easily
- RandomNSP: returns a random value from the selected terminology
- ConcatRandomNSP: will append a random value from the selected terminology to the input text (can be used mid route)
this node allow you to select 1 latent image from image batch for example if you generate 4 images, it allow you to select 1 of them to do further processing on it or you can use it to process them sequentially
This node is a variant of the original LatentUpscale tool, but instead of using width and height, you use a multiply number. For example, if the original image dimensions are (512,512) and the mul values are (2,2), the result image will be (1024,1024). You can also use it to downscale by using fractions, e.g., (512,512) mul (.5,.5) → (256,256).
This node is a variant of the original LatentUpscale tool, but instead of using width and height, you use a multiply number. For example, if the original image dimensions are (512,512) and the mul values are (2,2), the result image will be (1024,1024). You can also use it to downscale by using fractions, e.g., (512,512) mul (.5,.5) → (256,256).