GLM-OCR: Z.ai launches compact OCR model with CogViT and GLM-0.5B encoder-decoder

Z.ai has launched GLM-OCR, a compact, high-performance optical character recognition model built on its self-developed CogViT and GLM-0.5B encoder-decoder architecture.

The model uses a dedicated connection layer for cross-modal alignment and CLIP pre-training on billions of image-text pairs for visual semantic understanding and key token extraction. It is designed to stay lightweight for fast inference.