Scan Documents
with Mistral OCR

Cutting-edge optical character recognition (OCR) API designed to revolutionize document processing.

Unlock the OCR Power with Mistral

Revolutionize document processing with Mistral OCR – the cutting-edge Optical Character Recognition API that transforms complex documents into easy-to-read data and prompts.

Wanx 2.1

Advanced Scanning

Mistal can accurately extract text from complex PDFs, images, tables, mathematical formulas, and multilingual documents, capable of processing up to 2,000 pages per minute on a single node.

Structured Output

Unlike traditional OCR solutions, Mistral OCR maintains the formatting elements of documents, such as headings, paragraphs, and tables, ensuring extracted text is more valuable for subsequent applications

Doc-as-Prompt

Users can treat documents as prompts for more precise instructions, enabling the extraction of specific information and formatting it into structured outputs.

Mistral OCR Powerful Features

Mistral OCR offers advanced scanning capabilities for various document elements, including text, tables, mathematical formulas, and images.

Multiple Languages

Mistral OCR supports a wide range of languages and scripts, making it suitable for processing multilingual documents. It achieves high accuracy in text extraction, outperforming competitors like Google Document AI and Azure OCR.

Enhanced Reasoning
Audio ASR Performance

Math

Mistral OCR is particularly adept at recognizing and extracting mathematical equations from documents, including those formatted in LaTeX. It achieves a high accuracy rate in math recognition, making it ideal for scientific and academic applications.

Tables

Mistral OCR can accurately extract data from complex tables, including those with merged cells and nested structures. It boasts a high accuracy rate for table recognition, with benchmark scores showing it surpasses other leading models.

Audio Translation Performance
Audio ASR Performance

Document Scanning

Mistral OCR can process both text and images within documents, including interleaved imagery and complex layouts. It supports various document formats such as PDFs, images, and uploaded documents, making it versatile for different use cases.

Real-World Usage of Mistral OCR

Mistral empowers organizations across research, business, and legal sectors to unlock the full potential of their documents.

Speak with ChatGPT
Scientific Research

Convert scientific papers into AI-ready formats to enhance research efficiency, accelerate collaboration, and streamline workflows.

Chat about images
Legal Compliance

Securely process sensitive documents on-premises, extracting key information while ensuring regulatory compliance, faster legal research and review.

Real-time translation
Business Optimization 

Extract structured data from contracts, reports, and invoices to enable automation, optimize operations, and accelerate AI-driven decisions.

AI/ML API

Why Choose AI/ML API solution?

AI/ML API  provides scalability, faster deployment, and access to 200+ advanced machine learning models without the need for extensive in-house expertise or infrastructure.

Mixtral icon

Easy To Use

Our API allows seamless integration of powerful AI capabilities into your applications, regardless of your coding experience. Simply swap your API key to begin using the AI/ML API.

Google Icon

Scalable

AI/ML API provides flexibility for business growth since you can scale resources by purchasing more tokens as needed, ensuring optimal performance and cost efficiency

OpenAI Icon

Affordable

We offer flat, predictable pricing, payable by card or cryptocurrency, keeping it the lowest on the market and affordable for everyone.

import requests


def main():
    response = requests.post(
        "https://api.aimlapi.com/v1/ocr",
        headers={
            "Authorization": "Bearer <YOUR_API_KEY>",
            "Content-Type": "application/json",
        },
        json={
            "document": {
                "type": "document_url",
                "document_url": "https://css4.pub/2015/textbook/somatosensory.pdf"
            },
            "model": "mistral/mistral-ocr-latest",
        },
    )

    response.raise_for_status()
    data = response.json()

    print(data)


if __name__ == "__main__":
    main()

Getting started with
Mistral OCR API

AI Playground to quickly try Mistral OCR API.

For more information about technical features, please refer to the Mistral OCR model card.

Ready to get started? Get Your API Key Now!

Get API Key