GLMImage

GLMImage is a web-based AI image generation and editing tool designed for creators and professionals. You can create images by typing a detailed prompt, or upload a photo and iterate with image-to-image editing. It also supports a range of focused “scenes” (workflows) such as portraits, book covers, scientific illustrations, product-style visuals, and more.

Core capabilities include text-to-image generation, image-to-image editing with reference images, and bilingual text rendering for legible Chinese and English text embedded into generated graphics. You can also generate multiple image variants in a single request (up to 4) to compare directions without re-entering the same prompt. GLMImage records which model and provider were used for each task, and includes workflow and task records aimed at being auditable.

GLMImage is for anyone who wants to turn concepts into polished visuals quickly—especially when you need readable typography inside posters, covers, and marketing materials, or when you want to refine results by iterating on a reference image. Model weights are available via Hugging Face, and GLMImage may route requests to supported production models while treating the model stored for each task as authoritative.

Pricing-wise, new accounts receive 2 free images (2 credits) and can start generating immediately after signing in. After that, view pricing within GLMImage to purchase additional credits.

Do I need to sign in to generate images on GLMImage?

Yes—sign in is required to generate images. New accounts can start with 2 free images (2 credits).

What generation modes does GLMImage support?

GLMImage supports text-to-image creation from a prompt and image-to-image editing where you upload a reference image and iterate.

How do credits work on GLMImage?

GLMImage uses credits to generate images. New accounts receive 2 free images (2 credits), and you can view pricing for additional credits.

Does GLMImage add watermarks to generated images?

Generated images have no watermark by default.

Can GLMImage render text inside the generated images?

Yes. GLMImage can render legible Chinese and English text directly inside generated images, which is useful for posters, banners, and marketing graphics.

How can I compare multiple results for the same prompt?

Use batch generation to create up to 4 image variants in one request, then pick the direction you like best.