quwano/glm-ocr-desktop

GLM-OCR-Desktop is a desktop application that reads PDF, PNG, and JPG files and converts them into readable text. OCR stands for optical character recognition, the process of extracting text from images or scanned documents. This tool takes that a step further by also preserving document structure: when you choose Markdown output, headings and tables in the original document are converted into their Markdown equivalents, making the result useful for notes, editing, or feeding into other tools. The application handles two distinct situations. If a PDF already contains real embedded text (such as a PDF exported from a word processor), the tool extracts it directly without using an AI model, which is fast and accurate. If the input is a scanned image, an image only PDF, or a PNG or JPG file, the tool uses an AI vision model called GLM-OCR to analyze the layout and recognize the text. You can save the result as either Markdown (which preserves headings and tables) or plain unformatted text. The interface detects your system language automatically and displays in Japanese, English, or German. The application supports macOS and Windows, and an installer script is provided for both platforms that handles the setup of Python, the required libraries, and the AI model files. The AI model requires roughly 4 GB of free disk space. The tool is built in Python and released under the AGPL-3.0 open source license. You would use this when you need to turn scanned documents or image files into editable text, without uploading them to an online service.