The open-source image text-to-text model mistral-small-1 - Free deployment with support for multilingual processing

Mistral Small 1

Developed by CreitinGameplays

An image text-to-text model built on Mistral-Small-3.1-24B-Instruct-2503, supporting multilingual processing

Downloads 109

Release Time : 4/28/2025

Model Overview

This model focuses on converting text in images into editable text content, suitable for OCR and text processing tasks in multilingual scenarios

Multilingual support

Supports image text recognition and conversion in 24 languages

Large model architecture

Based on the Mistral architecture with 24B parameters, having strong text understanding ability

Instruction optimization

After instruction fine-tuning, it can better understand user needs and perform specific tasks

Image text recognition

Multilingual text conversion

Instruction understanding and execution

Large-scale text processing

Document digitization

Multilingual document scanning

Convert scanned multilingual documents into editable text

Improve document processing efficiency, supporting 24 languages

Business applications

Multilingual invoice processing

Automatically recognize and extract key information from multilingual invoices

Reduce manual entry and improve data processing accuracy

Property	Details
Model Type	Image-Text-to-Text
Base Model	mistralai/Mistral-Small-3.1-24B-Instruct-2503
Library Name	vllm
Supported Languages	English, French, German, Spanish, Portuguese, Italian, Japanese, Korean, Russian, Chinese, Arabic, Farsi, Indonesian, Malay, Nepali, Polish, Romanian, Serbian, Swedish, Turkish, Ukrainian, Vietnamese, Hindi, Bengali

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base