Skip to Content

Mistral OCR Tool

by
Odoo
v 18.0 Third Party 29
Download for v 18.0 Deploy on Odoo.sh
Availability
Odoo Online
Odoo.sh
On Premise
Odoo Apps Dependencies Discuss (mail)
Community Apps Dependencies
Lines of code 4652
Technical Name llm_tool_ocr_mistral
LicenseLGPL-3
Websitehttps://github.com/apexive/odoo-llm
Versions 16.0 18.0
You bought this module and need support? Click here!
Availability
Odoo Online
Odoo.sh
On Premise
Odoo Apps Dependencies Discuss (mail)
Community Apps Dependencies
Lines of code 4652
Technical Name llm_tool_ocr_mistral
LicenseLGPL-3
Websitehttps://github.com/apexive/odoo-llm
Versions 16.0 18.0

Read Any Document Like a Human—But 100x Faster

Extract text from invoices, receipts, handwritten notes, and scanned documents with AI-powered OCR

90% Time Saved
99.9% Accuracy
Zero Training

Is Your Team Drowning in Paper?

Every day, your team wastes hours manually retyping data from documents. Here's what that looks like:

2+ Hours Daily on Data Entry

Your accountant spends half their morning manually typing invoice data into Odoo. One typo means starting over.

Handwritten Notes Buried

Scanned receipts and handwritten vendor forms sit in folders because nobody wants to transcribe them.

Data Entry Errors Everywhere

Reading tiny text, switching between screens, manual typing = mistakes. Every error costs time and money to fix.

Can't Search Your Archives

You have boxes of old invoices and contracts, but they're just images. Finding anything requires digging through folders manually.

Before vs After

Before: Manual Entry

Step 1: Open scanned invoice PDF

Step 2: Read vendor name, squint at tiny text

Step 3: Type into Odoo (hoping no typos)

Step 4: Read invoice number, date, amount...

Step 5: Type each field manually

Step 6: Find mistake, fix it

Step 7: Repeat for line items...

15-20 minutes per document

After: AI-Powered OCR

Step 1: Upload document to Odoo

Step 2: Ask AI to extract text

Step 3: Review extracted data

Step 4: Done!

 

 

 

30 seconds per document

Your AI Document Reader—Available for Any LLM Assistant

This isn't just basic OCR. It's an AI-powered vision tool that understands your documents:

Reads invoices, receipts, bills, scanned documents, handwritten notes, product labels, screenshots—anything with text. Uses Mistral AI's vision models to extract text like a human would read it.

Works with any LLM assistant in your Odoo system. Once installed, all your AI assistants can read documents—no extra configuration needed.

Minimal dependencies, maximum power. No vector stores, no heavy infrastructure—just pure OCR capability ready to use.

Works beyond Odoo with MCP Server integration. Install our llm_mcp_server module and use this OCR tool from external AI clients like Cursor, Claude Desktop, and any MCP-compatible application. Parse PDFs and images in your Odoo database from anywhere—extract text from invoices, receipts, and documents without opening Odoo.

Real Time Savings

Let's say your team processes 30 documents per day (invoices, receipts, forms, etc.):

Manual Entry 15 min/doc × 30 docs 7.5 hours/day
With OCR Tool 30 sec/doc × 30 docs 15 minutes/day
Time Saved Daily 7.25 hours

That's 36+ hours saved per week

Almost a full-time employee's worth of work—freed up to focus on high-value tasks instead of data entry.

Powerful Features

Invoice & Receipt Parsing

Extract vendor names, dates, amounts, line items—everything you need from invoices and receipts.

Document Scanning

Turn scanned PDFs and images into searchable, editable text. Perfect for digitizing paper archives.

Handwriting Recognition

Extract text from handwritten notes, forms, and signatures. No more squinting at messy handwriting.

Multi-Page Support

Process documents with multiple pages automatically. Upload once, extract all text.

Minimal Dependencies

No knowledge base, no vector stores, no heavy infrastructure. Just pure OCR power.

LLM Assistant Ready

Works with any AI assistant in your system. Install once, use everywhere.

Real Use Cases

Automated Invoice Data Entry

Before: Accountant spends 2 hours entering 30 invoices manually.
After: AI extracts all data in 15 minutes. Accountant reviews and approves.

Digitizing Paper Archives

Before: Boxes of old contracts sit in storage, unsearchable.
After: Scan and extract text. Now you can search your entire archive instantly.

Receipt Expense Tracking

Before: Employees photograph receipts. Finance team squints and types.
After: AI reads receipts instantly. Expenses recorded in seconds.

Form Processing

Before: Customer fills paper form. Staff rekeys every field into Odoo.
After: Scan form, AI extracts data. Staff reviews and saves.

Product Label Extraction

Before: Inventory team manually enters product codes and descriptions.
After: Photograph label, AI reads everything. Product info auto-populated.

Screenshot Text Extraction

Before: Screenshots with error messages require manual transcription.
After: AI reads screenshot text instantly. Copy-paste ready.

How It Works

1
Install Module

Install this module in your Odoo instance

2
Configure Mistral

Add your Mistral API key and sync models

3
Use with Assistants

Any LLM assistant can now read documents

Extract Text

Upload documents and get instant text extraction

See It In Action

From setup to document extraction — here's how simple it is

1 Configure Mistral Provider

Add your Mistral API key and sync models. The provider setup is straightforward—just configure once and you're ready to use OCR across your entire Odoo system.

Mistral Provider Configuration

2 OCR Models Ready to Use

After syncing, Mistral's vision models appear automatically. These specialized OCR models (pixtral-12b or mistral-ocr-latest) are optimized for reading invoices, receipts, handwritten notes, and scanned documents with exceptional accuracy.

OCR models available for document parsing

3 Powerful Text Extraction in Action

Watch the magic happen: complex invoices, blurry scans, handwritten forms—the OCR engine reads them all and extracts clean, structured text. Even difficult-to-read documents are processed with near-perfect accuracy.

Before and after OCR text extraction comparison

Seamless Odoo Integration

This isn't a separate system. It's built into your Odoo workflow:

  • Works with existing LLM assistants — No new interface to learn
  • Attaches to any Odoo record — Invoices, vendors, projects, anywhere
  • Standard Odoo permissions — Tool consent managed through your existing settings
  • No external dependencies — Just needs Mistral AI provider configured
  • Instant setup — Install, configure API key, done

Trusted Technology

Developed by Apexive Solutions

Expert Odoo developers with deep LLM integration experience

Powered by Mistral AI

State-of-the-art vision models (mistral-ocr-latest) for accurate text extraction

Open Source

LGPL-3 license. View source on GitHub, contribute improvements

github.com/apexive/odoo-llm

Ready to Save 7+ Hours Every Day?

Stop manually typing data from documents. Let AI do the heavy lifting.

Install Now from Odoo Apps

Setup takes 5 minutes. See results immediately.

Mistral OCR Tool for Odoo 18.0

Developed by Apexive Solutions LLC

github.com/apexive/odoo-llm

License: LGPL-3

Mistral OCR Tool

Extract text from images and PDFs using Mistral AI vision models. Standalone OCR tool that any LLM assistant can use.

Module Type: 🔌 Extension (OCR Tool)

Architecture

┌───────────────────────────────────────────────────────────────┐
│                    AI Consumers                               │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────────────┐   │
│  │llm_assistant│  │  llm_letta  │  │   llm_mcp_server    │   │
│  └──────┬──────┘  └──────┬──────┘  └──────────┬──────────┘   │
└─────────┼────────────────┼────────────────────┼──────────────┘
          └────────────────┼────────────────────┘
                           ▼
              ┌───────────────────────────────────────────┐
              │   ★ llm_tool_ocr_mistral (This Module) ★  │
              │      Extract Text from Attachment Tool    │
              │  👁️ Vision OCR │ 📄 PDF │ 🖼️ Images       │
              └─────────────────────┬─────────────────────┘
                                    │
                        ┌───────────┴───────────┐
                        ▼                       ▼
    ┌───────────────────────────┐   ┌───────────────────────────┐
    │         llm_tool          │   │        llm_mistral        │
    │    (Tool Framework)       │   │    (Mistral Provider)     │
    └───────────────────────────┘   └───────────────────────────┘

Installation

What to Install

For OCR tool access:

odoo-bin -d your_db -i llm_tool_ocr_mistral

Auto-Installed Dependencies

  • llm (core infrastructure)
  • llm_tool (tool framework)
  • llm_mistral (Mistral AI provider)

Why Use This Module?

Feature llm_tool_ocr_mistral
OCR 👁️ Mistral vision models
Formats 📄 PDF, PNG, JPG, WEBP
Standalone 🔧 Tool for any assistant
MCP Ready 🔌 Works with external clients

Common Setups

I want to... Install
Invoice OCR llm_assistant_account_invoice (includes this)
Chat + OCR llm_assistant + llm_openai + llm_tool_ocr_mistral
Claude + OCR llm_mcp_server + llm_tool_ocr_mistral

What is OCR?

Optical Character Recognition (OCR) converts images and scanned documents into machine-readable text. This module provides an LLM tool that:

  1. Reads invoices, receipts, bills, and scanned documents
  2. Extracts text from handwritten notes and product labels
  3. Parses screenshots, photos, and multi-page PDFs

Unlike traditional OCR, Mistral's vision models understand document structure and context, providing accurate text extraction even from low-quality or handwritten sources.

Installation

Prerequisites

  • Odoo: 18.0 or higher
  • Required Modules: llm, llm_mistral, llm_tool
  • Mistral API Key: Get one from Mistral AI Platform

Install Steps

  1. Install required Odoo modules:

    odoo-bin -d your_database -i llm,llm_mistral,llm_tool,llm_tool_ocr_mistral

  2. Configure Mistral Provider:

    • Navigate to LLM → Providers → Mistral
    • Add your Mistral API key
    • Click Sync Models

  3. Restart Odoo server

Quick Start

1. Verify OCR Tool is Available

After installation, verify the tool is registered:

  • Navigate to LLM → Tools
  • Look for Extract Text from Attachment (OCR)
  • Status should be Active

2. Use OCR Tool

The tool is simple - give it an attachment ID, get back parsed text:

# Example: Extract text from attachment ID 5
thread = env['llm.thread'].create({})
thread.message_post(body="Please parse attachment ID 5")
thread.generate_response()

# Assistant uses extract_text_from_attachment tool
# Returns extracted text from the PDF/image

How It Works

Tool Interface

Tool Name: extract_text_from_attachment

Input:

{
  "attachment_id": 5
}

Output:

{
  "text": "Extracted text content from the document...",
  "filename": "invoice.pdf"
}

Usage Pattern

  1. LLM assistant needs to extract text from a document
  2. Assistant knows the attachment ID (from prompt, user message, or search)
  3. Assistant calls extract_text_from_attachment(attachment_id=5)
  4. Tool uses Mistral vision model to parse the attachment
  5. Returns clean, structured text

Usage Examples

Basic OCR

# Create thread and ask to parse attachment
thread = env['llm.thread'].create({})
thread.message_post(body="Parse attachment ID 123 and tell me the vendor name")
thread.generate_response()

Invoice Processing

# Invoice assistant with OCR
thread = env['llm.thread'].create({
    'model': 'account.move',
    'res_id': 456,
})
thread.message_post(body="Extract data from attachment ID 789")
thread.generate_response()

With llm_assistant_account_invoice Module

The invoice assistant module uses this OCR tool automatically:

# Install invoice assistant (includes OCR)
odoo-bin -d your_database -i llm_assistant_account_invoice

The assistant's prompt template tells it about attachments and how to use OCR.

Screenshots

1. Configure Mistral Provider

Mistral Provider

Add API key and sync models

2. OCR Models Available

OCR Models

Mistral vision models ready for parsing

3. Text Extraction

OCR Results

Clean text extracted from invoices and receipts

Supported Document Types

  • Invoices and Receipts: Vendor, date, amount, line items
  • Scanned Documents: Paper archives to searchable text
  • Handwritten Notes: Forms, signatures, meeting notes
  • Product Labels: Codes, descriptions, barcodes
  • Screenshots: Error messages, UI text
  • Multi-page PDFs: All pages processed automatically

Use with External Clients (MCP)

Install llm_mcp_server to expose this tool to external AI clients:

odoo-bin -d your_database -i llm_mcp_server

Compatible Clients:

  • Claude Desktop
  • Cursor IDE
  • Claude Code CLI
  • Any MCP-compatible application

Example from Claude Desktop:

You: "Parse attachment ID 123 from my Odoo database"
Claude: [Uses OCR tool via MCP] "The document contains..."

This lets you extract text from Odoo attachments without opening Odoo UI.

Configuration

Troubleshooting

Tool not found?

  • Check llm_tool_ocr_mistral is installed
  • Verify tool is active in LLM → Tools

No vision models?

  • Go to LLM → Providers → Mistral
  • Click Sync Models
  • Verify API key is valid

Extraction fails?

  • Verify attachment ID exists
  • Check attachment is image/PDF format
  • Ensure Mistral vision model is available

Poor quality results?

  • Use higher resolution images (300+ DPI)
  • Ensure text is legible
  • Try clearer scan/photo

Documentation

  • GitHub: https://github.com/apexive/odoo-llm

License

LGPL-3 - https://www.gnu.org/licenses/lgpl-3.0.html

---

© 2025 Apexive Solutions LLC

Please log in to comment on this module

  • The author can leave a single reply to each comment.
  • This section is meant to ask simple questions or leave a rating. Every report of a problem experienced while using the module should be addressed to the author directly (refer to the following point).
  • If you want to start a discussion with the author, please use the developer contact information. They can usually be found in the description.
Please choose a rating from 1 to 5 for this module.