ArmorOCR-GGUF

About

GGUF-quantized builds of inclusionAI's ArmorOCR, an image-text-to-text vision-language model for OCR built on Qwen3-VL and tagged for adversarial and grounded OCR in English and Chinese. The GGUF format lets the model run on local hardware through llama.cpp, Ollama, LM Studio, or Docker Model Runner, so text extraction can happen without sending images to a hosted API. It is released under Apache-2.0.