PixelFlow Components

Building Blocks of PixelFlow

Canvas: The canvas in PixelFlow is like a white board. This is where you create your workflows. You can simply drag and drop the nodes onto the canvas and string them together.


Input Nodes (Image, Audio, Text): These are the nodes for your inputs. Inputs can be in the form of image, text, or audio depending on the models you are using in your workflows. You can select the appropriate input node based on the type of data you want to process.

Output Nodes (Image, Audio, Text, Video): These are the nodes for your outputs. Outputs can be image, text, video, or audio depending on the models you are using in your workflows. After processing the input data through your workflow, the results will be delivered through these output nodes.

Model Nodes:

  1. Models: These are the model APIs you will be using in your workflows. They are connected with input nodes and output nodes. You can find a variety of models such as txt2img, img2img, Controlnets, Inpainting models, and audio & video models. These models are the core AI components of your workflows that are responsible for processing the input data and generating the output.

  2. Search Models: To quickly access the models, you can use the search function. This allows you to search for any particular model that is currently available on the Segmind model list. This feature makes it easy to find and add the right models to your workflow.

*The key to creating effective workflows in PixelFlow is understanding the type of data you’re working with (image, text, audio, video) and choosing the appropriate input and output nodes. This allows you to fully leverage the capabilities of the models in your workflow.

Node connections:

Connections play a crucial role in PixelFlow as they link different nodes on the canvas, forming the structure of your workflows. Port: A port is an edge on a node where connections are made with other nodes. Think of it as a docking point where data can flow in or out, connecting one node to another. By understanding and effectively using ports, you can ensure smooth data flow and create more complex and powerful workflows.

  • Input and Output Connections: An input node can only take a single connection line, meaning it can receive data from one source at a time. On the other hand, an output node can create multiple connection lines, allowing it to send data to multiple nodes simultaneously. This enables the creation of complex workflows where one node’s output serves as the input for multiple nodes.

Run Button:

The ‘Run’ button executes all workflows on the canvas. Be careful when using this option, especially if you have multiple workflows. It’s best to use it only if you want to run all workflows together. Remember, the time it takes to get outputs can vary based on how complex each workflow is (i.e., the number of nodes) and the total number of workflows. You can refer to the ‘Executing Nodes’ for more information.

Last updated