Ideogram P-Image - fast text-to-image

Generate images from text descriptions using Ideogram’s fast P-Image model in ComfyUI, with quality comparable to leading models at lower cost and latency

Ideogram P-Image is Ideogram’s fast text-to-image model, delivering quality comparable to leading image models at significantly lower cost and latency. It accepts a single text prompt and outputs one image, with controls for quality level and resolution, making it ideal for rapid iteration and high-volume generation.

What Ideogram P-Image is good at

• Speed and value: Quality comparable to leading image models at lower cost and latency
• Quality tiers: Choose from VERY_LOW, LOW, MEDIUM, or HIGH speed/price/quality tiers
• Flexible output: 1K or 2K size classes with full aspect ratio support
• Prompt upsampling: AUTO expands short prompts into detailed structured captions before generation

Example outputs

Text-to-image generation from a single prompt:

Use it in ComfyUI

Ideogram P-Image workflows

Run the text-to-image workflow in ComfyUI, locally or on Comfy Cloud