Why Data Alchemy uses Claude AI (and why it will soon have its own LLM)

In Data Alchemy the LLM is the engine that reads and understands every document. The quality of that model determines the quality of the extraction. Here's how we choose the LLM, why we use Claude AI by Anthropic today, and where we're heading.

A dedicated LLM for each document model

Data Alchemy doesn't orchestrate several models on the same document. It works differently, and more robustly: for each document model you create — invoices, delivery notes, orders, or any other type — the system pairs one LLM that handles all documents of that model.

This approach has concrete advantages:

• >Specialization: each document type is handled by the model best suited to that specific task.
• >Consistency: all documents of the same model are read with the same logic, reducing variation in results.
• >Control: you know exactly which LLM processes each document type.

In other words, we don't pick a model "in general": we assign the right LLM to each document type.

Today: Claude AI, the best model for documents and images

For extracting data from documents and images, today the best model on the market is Anthropic's Claude AI. We've verified it in the field, on real business documents:

• >Superior accuracy in data extraction compared to other models.
• >Instruction following: it returns exactly the requested fields, in the requested format.
• >Low tendency to hallucinate: when a value isn't there, it flags it instead of inventing one.
• >Understanding of documents and images: it reads scans, photos and complex layouts well.

That's why Data Alchemy today uses Claude AI as its main engine and recommends it to its clients: when it comes to understanding a document, it's the best tool available.

Tomorrow: Data Alchemy's own LLM

Using the best model on the market is the right choice today, but our goal is to go further. By the end of the third quarter (Q3), Data Alchemy will have its own LLM, developed in-house and specialized in document data extraction. In most cases, it will replace Claude AI.

The reasons behind this choice are strategic:

• >Data sovereignty: a proprietary model further strengthens control and keeps data within Europe.
• >Specialization: an LLM built specifically for business documents, not a generalist model.
• >Independence: less reliance on third-party providers, more control over cost, performance and product evolution.

Claude AI will still be available where it's the best fit: our architecture lets us assign the most suitable LLM — proprietary or third-party — to each document model.

What it means for your documents

For you, little changes in day-to-day use and a lot changes in substance: for every document type, you always get the engine best suited to read it. Today that's Claude AI; with the arrival of our own model, you'll get the same quality with even more control and data sovereignty.

---

Want to see the extraction quality on your own documents? Book a free demo and try it with Data Alchemy's AI engine.