Build RAG Chatbot with Haystack, Haystack In-memory store, STACKIT Mistral-Nemo-Instruct-2407-FP8, and Ollama paraphrase-multilingual

Introduction to RAG

Retrieval-Augmented Generation (RAG) is a game-changer for GenAI applications, especially in conversational AI. It combines the power of pre-trained large language models (LLMs) like OpenAI’s GPT with external knowledge sources stored in vector databases such as Milvus and Zilliz Cloud, allowing for more accurate, contextually relevant, and up-to-date response generation. A RAG pipeline usually consists of four basic components: a vector database, an embedding model, an LLM, and a framework.

Key Components We'll Use for This RAG Chatbot

This tutorial shows you how to build a simple RAG chatbot in Python using the following components:

Haystack: An open-source Python framework designed for building production-ready NLP applications, particularly question answering and semantic search systems. Haystack excels at retrieving information from large document collections through its modular architecture that combines retrieval and reader components. Ideal for developers creating search applications, chatbots, and knowledge management systems that require efficient document processing and accurate information extraction from unstructured text.
Haystack in-memory store: a very simple, in-memory document store with no extra services or dependencies. It is great for experimenting with Haystack, and we do not recommend using it for production. If you want a much more scalable solution for your apps or even enterprise projects, we recommend using Zilliz Cloud, which is a fully managed vector database service built on the open-source Milvusand offers a free tier supporting up to 1 million vectors.)
STACKIT Mistral-Nemo-Instruct-2407-FP8: An instruction-tuned AI model optimized for efficient, real-time natural language processing. Combining Mistral’s robust language capabilities with Nemo’s scalability, it excels in low-latency text generation, query resolution, and task automation using FP8 precision. Ideal for resource-constrained environments, customer support automation, and edge-computing applications requiring rapid, accurate responses.
Ollama Paraphrase-Multilingual: A versatile AI model designed to rephrase and restructure text across multiple languages while preserving meaning. Strengths include multilingual adaptability, context retention, and semantic accuracy. Ideal for translation enhancement, cross-lingual content generation, global customer support, and academic or technical writing requiring nuanced paraphrasing in diverse linguistic contexts.

By the end of this tutorial, you’ll have a functional chatbot capable of answering questions based on a custom knowledge base.

Note: Since we may use proprietary models in our tutorials, make sure you have the required API key beforehand.

Step 1: Install and Set Up Haystack

import os
import requests

from haystack import Pipeline
from haystack.components.converters import MarkdownToDocument

from haystack.components.preprocessors import DocumentSplitter
from haystack.components.writers import DocumentWriter

Step 2: Install and Set Up STACKIT Mistral-Nemo-Instruct-2407-FP8

STACKIT is the cloud and colocation provider of the Schwarz Group. We can use different models on its cloud services with ease through its API.

pip install stackit-haystack

from haystack_integrations.components.generators.stackit import STACKITChatGenerator
from haystack.dataclasses import ChatMessage

generator = STACKITChatGenerator(model="neuralmagic/Mistral-Nemo-Instruct-2407-FP8")

Step 3: Install and Set Up Ollama paraphrase-multilingual

pip install ollama-haystack

Make sure that you have a running Ollama model (either through a docker container, or locally hosted). No other configuration is necessary as Ollama has the embedding API built in.

from haystack import Document
from haystack_integrations.components.embedders.ollama import OllamaDocumentEmbedder
from haystack_integrations.components.embedders.ollama import OllamaTextEmbedder

text_embedder = OllamaTextEmbedder(model="paraphrase-multilingual")
document_embedder = OllamaDocumentEmbedder(model="paraphrase-multilingual")

Step 4: Install and Set Up Haystack In-memory store

from haystack.document_stores.in_memory import InMemoryDocumentStore
from haystack.components.retrievers import InMemoryEmbeddingRetriever

document_store = InMemoryDocumentStore()
retriever=InMemoryEmbeddingRetriever(document_store=document_store))

Step 5: Build a RAG Chatbot

Now that you’ve set up all components, let’s start to build a simple chatbot. We’ll use the Milvus introduction doc as a private knowledge base. You can replace it your own dataset to customize your RAG chatbot.

url = 'https://raw.githubusercontent.com/milvus-io/milvus-docs/refs/heads/v2.5.x/site/en/about/overview.md'
example_file = 'example_file.md'
response = requests.get(url)
with open(example_file, 'wb') as f:
    f.write(response.content)
file_paths = [example_file]  # You can replace it with your own file paths.

indexing_pipeline = Pipeline()
indexing_pipeline.add_component("converter", MarkdownToDocument())
indexing_pipeline.add_component("splitter", DocumentSplitter(split_by="sentence", split_length=2))
indexing_pipeline.add_component("embedder", document_embedder)
indexing_pipeline.add_component("writer", DocumentWriter(document_store))
indexing_pipeline.connect("converter", "splitter")
indexing_pipeline.connect("splitter", "embedder")
indexing_pipeline.connect("embedder", "writer")
indexing_pipeline.run({"converter": {"sources": file_paths}})

# print("Number of documents:", document_store.count_documents())
  
question = "What is Milvus?"  # You can replace it with your own question.

retrieval_pipeline = Pipeline()
retrieval_pipeline.add_component("embedder", text_embedder)
retrieval_pipeline.add_component("retriever", retriever)
retrieval_pipeline.connect("embedder", "retriever")

retrieval_results = retrieval_pipeline.run({"embedder": {"text": question}})

# for doc in retrieval_results["retriever"]["documents"]:
#     print(doc.content)
#     print("-" * 10)

from haystack.utils import Secret
from haystack.components.builders import PromptBuilder

retriever=InMemoryEmbeddingRetriever(document_store=document_store) 
 text_embedder = OllamaTextEmbedder(model="paraphrase-multilingual")

prompt_template = """Answer the following query based on the provided context. If the context does
                     not include an answer, reply with 'I don't know'.\n
                     Query: {{query}}
                     Documents:
                     {% for doc in documents %}
                        {{ doc.content }}
                     {% endfor %}
                     Answer: 
                  """

rag_pipeline = Pipeline()
rag_pipeline.add_component("text_embedder", text_embedder)
rag_pipeline.add_component("retriever", retriever)
rag_pipeline.add_component("prompt_builder", PromptBuilder(template=prompt_template))
rag_pipeline.add_component("generator", generator)
rag_pipeline.connect("text_embedder.embedding", "retriever.query_embedding")
rag_pipeline.connect("retriever.documents", "prompt_builder.documents")
rag_pipeline.connect("prompt_builder", "generator")

results = rag_pipeline.run({"text_embedder": {"text": question}, "prompt_builder": {"query": question},})
print('RAG answer:\n', results["generator"]["replies"][0])

Optimization Tips

As you build your RAG system, optimization is key to ensuring peak performance and efficiency. While setting up the components is an essential first step, fine-tuning each one will help you create a solution that works even better and scales seamlessly. In this section, we’ll share some practical tips for optimizing all these components, giving you the edge to build smarter, faster, and more responsive RAG applications.

Haystack optimization tips

To optimize Haystack in a RAG setup, ensure you use an efficient retriever like FAISS or Milvus for scalable and fast similarity searches. Fine-tune your document store settings, such as indexing strategies and storage backends, to balance speed and accuracy. Use batch processing for embedding generation to reduce latency and optimize API calls. Leverage Haystack's pipeline caching to avoid redundant computations, especially for frequently queried documents. Tune your reader model by selecting a lightweight yet accurate transformer-based model like DistilBERT to speed up response times. Implement query rewriting or filtering techniques to enhance retrieval quality, ensuring the most relevant documents are retrieved for generation. Finally, monitor system performance with Haystack’s built-in evaluation tools to iteratively refine your setup based on real-world query performance.

Haystack in-memory store optimization tips

Haystack in-memory store is just a very simple, in-memory document store with no extra services or dependencies. We recommend that you just experiment it with RAG pipeline within your Haystack framework, and we do not recommend using it for production. If you want a much more scalable solution for your apps or even enterprise projects, we recommend using Zilliz Cloud, which is a fully managed vector database service built on the open-source Milvusand offers a free tier supporting up to 1 million vectors

STACKIT Mistral-Nemo-Instruct-2407-FP8 optimization tips

To optimize STACKIT Mistral-Nemo-Instruct-2407-FP8 in a RAG setup, ensure input prompts are concise and contextually enriched with retrieved documents, truncating irrelevant sections to stay within the 4k token limit. Leverage FP8 precision for faster inference by enabling compatible hardware acceleration (e.g., NVIDIA Tensor Cores). Batch process queries when possible, and fine-tune retrieval thresholds to balance relevance and noise. Use caching for frequent queries, and monitor latency to adjust chunk sizes or parallelize document processing. Regularly validate outputs against ground truth to refine retrieval-model alignment.

Ollama paraphrase-multilingual optimization tips

To optimize Ollama paraphrase-multilingual in a RAG setup, preprocess input text to remove noise and standardize formats (e.g., lowercasing, punctuation normalization). Use smaller temperature values (e.g., 0.3) for deterministic outputs and adjust max_length to balance context retention and brevity. Batch processing parallelizes paraphrasing for efficiency. Cache frequent or repetitive queries to reduce redundant computations. Validate outputs with metrics like BLEU or semantic similarity scores. For multilingual use, explicitly specify language codes in prompts to avoid ambiguity. Fine-tune on domain-specific data if available, and leverage GPU acceleration for faster inference.

By implementing these tips across your components, you'll be able to enhance the performance and functionality of your RAG system, ensuring it’s optimized for both speed and accuracy. Keep testing, iterating, and refining your setup to stay ahead in the ever-evolving world of AI development.

RAG Cost Calculator: A Free Tool to Calculate Your Cost in Seconds

Estimating the cost of a Retrieval-Augmented Generation (RAG) pipeline involves analyzing expenses across vector storage, compute resources, and API usage. Key cost drivers include vector database queries, embedding generation, and LLM inference.

RAG Cost Calculator is a free tool that quickly estimates the cost of building a RAG pipeline, including chunking, embedding, vector storage/search, and LLM generation. It also helps you identify cost-saving opportunities and achieve up to 10x cost reduction on vector databases with the serverless option.

Calculate your RAG cost now.

Calculate your RAG cost

What Have You Learned?

By now, you’ve unlocked the magic of building a RAG system from the ground up! You learned how Haystack acts as the backbone, seamlessly stitching together every component. The Haystack In-memory store became your go-to vector database, handling lightning-fast similarity searches without breaking a sweat. Together with Ollama’s paraphrase-multilingual embedding model, you transformed raw text into rich numerical representations, bridging the gap between human language and machine understanding. Then came the star of the show—STACKIT Mistral-Nemo-Instruct-2407-FP8, the LLM that turned those retrieved snippets into coherent, context-aware answers. You saw firsthand how each piece plays a critical role: the framework orchestrates, the database retrieves, the embeddings encode, and the LLM generates. It’s like watching a symphony where every instrument harmonizes perfectly!

But wait—there’s more! You also picked up pro tips for optimizing your pipeline, like tweaking chunk sizes for better performance and balancing speed with accuracy. And let’s not forget the free RAG cost calculator, your secret weapon for estimating expenses before scaling up. Imagine the possibilities now: custom chatbots, dynamic Q&A systems, or even AI-powered research tools. The world of RAG is yours to explore! So, what’s next? Start experimenting, apply those optimizations, and let your creativity run wild. Build something that wows users, solves real problems, or pushes the boundaries of what AI can do. You’ve got the tools, the know-how, and the spark—now go light up the world with your innovations! 🚀

Further Resources

🌟 In addition to this RAG tutorial, unleash your full potential with these incredible resources to level up your RAG skills.

How to Build a Multimodal RAG | Documentation
How to Enhance the Performance of Your RAG Pipeline
Graph RAG with Milvus | Documentation
How to Evaluate RAG Applications - Zilliz Learn
Generative AI Resource Hub | Zilliz

We'd Love to Hear What You Think!

We’d love to hear your thoughts! 🌟 Leave your questions or comments below or join our vibrant Milvus Discord community to share your experiences, ask questions, or connect with thousands of AI enthusiasts. Your journey matters to us!

If you like this tutorial, show your support by giving our Milvus GitHub repo a star ⭐—it means the world to us and inspires us to keep creating! 💖