An open-source, minimal Python wrapper around the Mistral API focused on OCR use cases where visual structure matters—such as tables, charts, and diagrams. The wrapper produces markdown output with image clippings embedded directly, making it suitable for downstream documentation and knowledge pipelines.

Links

Project Repository | Medium Article

Key Features

• Designed for clarity, minimal surface area, and ease of integration.
• Minimal wrapper that abstracts away SDK setup and boilerplate.
• Focused on OCR scenarios where images must be preserved alongside text.
• Embeds extracted images directly into the markdown output as Base64.
• Exposes a deliberately small API surface with only two primary entry points.

1# OCR an image
2await mistral_client.ocr_image("path/to/image.png")
3
4# OCR a PDF
5await mistral_client.ocr_pdf("path/to/document.pdf")