GLM Image

GLM Image generates and edits pictures from text prompts. It is an AI image generator and photo editor built on GLM-Image diffusion technology, and it also handles image to image work.

Output goes up to 2K resolution, and the site says an optimized VAE pipeline with Smart Compression keeps generation quick.

Prompts can be written in Chinese or English, which is handy if you switch between the two.

GLM Image also refines your prompt automatically before generating, so a rough one-line idea still gets expanded.

Training with RLHF is meant to push results toward images people find more appealing, and a dual-encoder setup with Semantic Hierarchy is credited with following complex prompts closely.

Several aspect ratios are supported, so you can match the format of different platforms.

The pricing model is free, but you need to log in before you can generate anything.

The catch is commercial use: it is limited to the paid Pro and Max plans.

If you plan to sell or publish the images for business, expect to pay.