How to use Deepset AI

Published On - March 10, 2025

Jonathan Dough Content

Artificial Intelligence is exciting. But it can also feel overwhelming. That’s where Deepset AI comes in. It helps you work with AI models in a simple and effective way.

In this guide, you’ll learn how to use Deepset AI, step by step. We’ll keep things fun and easy!

What is Deepset AI?

Deepset AI is a platform for working with large language models (LLMs). It helps you build AI-powered applications for tasks like:

  • Answering questions
  • Searching documents
  • Extracting information
  • Summarizing text

It is especially useful for businesses and researchers who want to use AI without extra complexity.

Getting Started

Before you begin, you need to set up your environment. Here’s what to do:

  1. Install Python: Deepset AI works with Python. If you don’t have it, download and install it first.
  2. Install Haystack: Haystack is Deepset AI’s core framework. You can install it with this command:
pip install farm-haystack

Now you’re ready to start using Deepset AI!

How to Use Deepset AI

Step 1: Load a Language Model

To work with Deepset AI, you need a language model. Here’s how to load one:

from haystack.nodes import TransformersReader
reader = TransformersReader(model_name="deepset/roberta-base-squad2")

This loads a model that can answer questions based on text input.

Step 2: Add a Document Store

Next, you need a place to store documents the AI will search:

from haystack.document_stores import InMemoryDocumentStore
document_store = InMemoryDocumentStore()

There are other options, like Elasticsearch and FAISS, but this one is simple and great for testing.

Step 3: Add Documents

Now, let’s add some documents for the AI to read:

docs = [{"content": "The Eiffel Tower is in Paris, France.", "meta": {"source": "Wikipedia"}}]
document_store.write_documents(docs)

With this, the AI now has knowledge to answer questions!

Step 4: Ask a Question

Time to ask a question and get an answer:

prediction = reader.predict(question="Where is the Eiffel Tower?", documents=docs)
print(prediction)

AI will respond with: “Paris, France.” Amazing, right?

Why Use Deepset AI?

Here are some key reasons:

  • It’s powerful: It can process large amounts of text quickly.
  • It’s flexible: Supports different AI models and databases.
  • It’s easy to use: Simple setup, great documentation.

More Cool Features

Deepset AI doesn’t stop at simple Q&A. You can also:

  • Use retrievers to search large datasets efficiently.
  • Apply summarization models to shorten long documents.
  • Fine-tune models for better accuracy on your own data.

Where to Go Next?

If you want to dive deeper, check out:

Have fun experimenting!

Learning