ComfyUI: Text-to-image generation with PiD DIT model

ComfyUI Try it →Watch full video →AI Search · Jun 2026

Demo summary

The creator demonstrates the standalone PiD text-to-image workflow, generating an image of a leopard in a jungle directly from a prompt using the lightweight DIT model.

Step-by-step

Drag and drop the workflow file onto the ComfyUI interface
Download the text-to-image model from the provided link under files and versions
Place the downloaded model file into the 'models/diffusion_models' directory
Press 'R' in ComfyUI to refresh the model list
Select the downloaded model from the dropdown menu
Select the 'Gemma 2 2b' model for the clip encoder
Enter your prompt and set the desired width, height, and batch size
Press 'Run' to generate the image

Options

Use the BF-16 model variant
Use the smaller MXFP8 model variant if using a Blackwell architecture or 50 series Nvidia GPU
Adjust batch size to generate multiple images at once

Watch out for

The Gemma 2 2b clip encoder must be previously downloaded to be selectable

Tips

Note that this model is highly efficient at only 2.6 GB compared to Z image or Flux 2

Highlights

“notice that this is blazing fast. So, it should be able to generate an image in like less than 10 seconds”

All demos from “The BEST AI for 4K images. Free & fast”

6:420:59Configure Gemma 2B Text Encoder in ComfyUIThe creator shows how to load the Gemma 2B text encoder model into the ComfyUI workflow and select it from the node dropdown menu.ComfyUI· AI Image Generator
9:521:53Upscale an existing image to 4K with PiDThe video demonstrates uploading a landscape photo into a ComfyUI workflow and using the PiD (Pixel Diffusion) model to upscale it to 4096 resolution in under 10 seconds.ComfyUI· AI Image Upscaler
13:200:38Compare original vs upscaled image using RG3 nodesThe user adds an 'image compare' node by RG3 to the ComfyUI canvas to show a side-by-side slider comparison of the blurry original image versus the sharp PiD upscaled result.ComfyUI· AI Image Upscaler
14:274:08Generate and upscale images using Zee Image and PiDA demonstration of a combined workflow where Zee Image Turbo generates a 1K image of a red panda which is then automatically passed into the PiD upscaler for a 4K final output.ComfyUI· AI Image Upscaler
20:151:27Text-to-image generation with PiD DIT modelCurrentThe creator demonstrates the standalone PiD text-to-image workflow, generating an image of a leopard in a jungle directly from a prompt using the lightweight DIT model.ComfyUI· Text to Image
Watch “The BEST AI for 4K images. Free & fast” →

Text to Image

7:390:44Basic text-to-image generation in ComfyUIThe user demonstrates a basic text-to-image workflow using the RealVisXL model to generate an image of a castle in a forest.AI Search
22:350:32Generating an image with Flux modelDemonstrates a text-to-image generation using the Flux model in ComfyUI, including prompt entry and resolving connection errors.Sebastian Kamph
20:151:27Text-to-image generation with PiD DIT modelCurrentThe creator demonstrates the standalone PiD text-to-image workflow, generating an image of a leopard in a jungle directly from a prompt using the lightweight DIT model.AI Search
4:250:43Generate images with Stable Diffusion 3.5The video demonstrates entering a text prompt, selecting CLIP models, and clicking 'Queue' to generate a high-definition image using Stable Diffusion 3.5.AIPURE
12:211:24Text-to-image generation with Z-Turbo and QwenDemonstration of the Z-Turbo model and Qwen text encoder to generate specific advertising text on the side of a 3D tram model within a ComfyUI workflow.Matt Hallett Visual
2:501:21Generate an AI image from a text promptThe video shows the process of entering positive and negative prompts, configuring image dimensions, and clicking 'Run' to generate a chocolate chip cookie image in ComfyUI.@KevinStratvert
7:020:23Text-to-image generation with FluxThe user demonstrates Flux's prompt adherence by generating an image of an old TV with the word 'flux' on it in an abandoned workshop.Artificial Images