GLM-Image
Zhipu's autoregressive image generation model — complex semantics + high-quality images
Zhipu's autoregressive image generation model that understands complex semantics and generates high-quality images.
Image generation and computer vision — detection, editing, 3D.
Key features
- Hybrid autoregressive (9B) plus diffusion decoder (7B DiT) image generation architecture
- Strong text rendering via Glyph Encoder text module; excels at knowledge-intensive
- Single model for text-to-image and image-to-image: editing, style transfer, identity
- Multi-subject consistency and identity-preserving generation for people and objects
- Implemented natively in transformers and diffusers libraries; API available
Good for
- Information-dense image generation (infographics, recipes, posters)
- Image editing and style transfer with identity preservation
Stack & tags
PythonPyTorchimage-generationtext-to-imageimage-editingtext-renderingdiffusionopen-weightszhipu
Quick start
pip install git+https://github.com/huggingface/transformers.git
pip install git+https://github.com/huggingface/diffusers.gitFrom the project README on GitHub.
Similar open-source projects
FLUX.2 klein
★ 25,996Black Forest Labs' sub-second image generation — generate and edit in <0.5s
DiffSynth-Studio
★ 13,200The framework behind Qwen-Image-i2L — the first "Image→LoRA" visual generation system