Skip to content

CogView4

A bilingual text to image diffusion model

Visit official site

What CogView4 is

The model is described as a diffusion based text to image generator from Tsinghua University supporting Chinese and English input.

What you can do with it

  • Generate images from Chinese or English text
  • Produce high resolution images
  • Use inference options such as BF16 precision

Application scenarios

  • Generate a sci fi scene for a film poster
  • Produce inspiration sketches for artists
  • Create teaching images for complex concepts

Pros & cons

✓ What we like

  • Diffusers and Gradio frameworks are supported
  • The model was shown at ECCV 2024

! What to watch out for

  • Named inference options centre on BF16 and given frameworks
  • Input is provided as a text prompt

FAQ

What resolution is listed?

Up to 2048x2048.

Which frameworks are named?

Diffusers and Gradio.

Last reviewed: 2026-11-15

More AI image generation tools

View all →

How we review