Build RAG Chatbot with Haystack, Milvus, Google Vertex AI Gemini 2.0 Flash, and HuggingFace all-mpnet-base-v2

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.
Milvus: An open-source vector database optimized to store, index, and search large-scale vector embeddings efficiently, perfect for use cases like RAG, semantic search, and recommender systems. If you hate to manage your own infrastructure, we recommend using Zilliz Cloud, which is a fully managed vector database service built on Milvus and offers a free tier supporting up to 1 million vectors.
Google Vertex AI Gemini 2.0 Flash: A lightweight, high-speed AI model optimized for rapid inference and cost-effective scalability. It excels in real-time applications requiring low latency, such as chatbots, summarization, and data processing, balancing performance with efficiency for high-volume enterprise workloads on Google Cloud.
HuggingFace all-mpnet-base-v2: A versatile sentence-transformers model optimized for generating high-quality semantic embeddings. Leveraging MPNet's masked and permuted pretraining, it excels in capturing nuanced text semantics, offering robust performance in multilingual and domain-specific tasks. Ideal for semantic search, text clustering, similarity comparison, and information retrieval due to its efficiency and accuracy.

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 Google Vertex AI Gemini 2.0 Flash

Using theVertexAIGeminiGenerator with Haystack requires authentication using Google Cloud Application Default Credentials (ADCs). This means your application must be set up with credentials that allow it to access Google Cloud services. If you're not sure how to configure ADCs, check the official Google documentation for setup instructions.

It's important to use a Google Cloud account that has the right permissions to access a project with Google Vertex AI endpoints. Without proper access, the generator won’t work as expected.

To find your project ID, you can either look it up in the Google Cloud Console under the resource manager or run the following command in your terminal.

Now let's install and set up this model.

pip install google-vertex-haystack

from haystack_integrations.components.generators.google_vertex import VertexAIGeminiGenerator

generator = VertexAIGeminiGenerator(model="gemini-2.0-flash-001")

Step 3: Install and Set Up HuggingFace all-mpnet-base-v2

Haystack'sHuggingFaceAPITextEmbedder can be used to embed strings with different Hugging Face APIs:

The component uses a HF_API_TOKEN environment variable by default. Otherwise, you can pass a Hugging Face API token at initialization with token – see code examples below. The token is needed:

If you use the Serverless Inference API, or
If you use Inference Endpoints.

Here, in this tutorial, we'll use the Free Serverless Inference API. Let's install and set up the model.

To use this API, you need a free Hugging Face token. The Embedder expects the model in api_params.

from haystack.components.embedders import HuggingFaceAPITextEmbedder
from haystack.utils import Secret
from haystack.components.embedders import HuggingFaceAPIDocumentEmbedder
from haystack.dataclasses import Document

text_embedder = HuggingFaceAPITextEmbedder(api_type="serverless_inference_api",
                                           api_params={"model": "sentence-transformers/all-mpnet-base-v2"},
                                           token=Secret.from_token("<your-api-key>"))
document_embedder = HuggingFaceAPIDocumentEmbedder(api_type="serverless_inference_api",
                                              api_params={"model": "sentence-transformers/all-mpnet-base-v2"},
                                              token=Secret.from_token("<your-api-key>"))

Step 4: Install and Set Up Milvus

pip install --upgrade pymilvus milvus-haystack

from milvus_haystack import MilvusDocumentStore
from milvus_haystack.milvus_embedding_retriever import MilvusEmbeddingRetriever

document_store = MilvusDocumentStore(connection_args={"uri": "./milvus.db"}, drop_old=True,)
retriever = MilvusEmbeddingRetriever(document_store=document_store, top_k=3)

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 = MilvusEmbeddingRetriever(document_store=document_store, top_k=3) 
 text_embedder = HuggingFaceAPITextEmbedder(api_type="serverless_inference_api",
                                           api_params={"model": "sentence-transformers/all-mpnet-base-v2"},
                                           token=Secret.from_token("<your-api-key>"))

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.

Milvus optimization tips

Milvus serves as a highly efficient vector database, critical for retrieval tasks in a RAG system. To optimize its performance, ensure that indexes are properly built to balance speed and accuracy; consider utilizing HNSW (Hierarchical Navigable Small World) for efficient nearest neighbor search where response time is crucial. Partitioning data based on usage patterns can enhance query performance and reduce load times, enabling better scalability. Regularly monitor and adjust cache settings based on query frequency to avoid latency during data retrieval. Employ batch processing for vector insertions, which can minimize database lock contention and enhance overall throughput. Additionally, fine-tune the model parameters by experimenting with the dimensionality of the vectors; higher dimensions can improve retrieval accuracy but may increase search time, necessitating a balance tailored to your specific use case and hardware infrastructure.

Google Vertex AI Gemini 2.0 Flash optimization tips

To optimize Gemini 2.0 Flash in RAG, balance chunk size for retrieved documents—smaller chunks (200-400 tokens) improve precision, while larger ones preserve context. Use preprocessing to clean text, remove noise, and add metadata (e.g., timestamps) for relevance scoring. Adjust model parameters: lower temperature (0.2-0.5) for factual outputs, and limit max_output_tokens to reduce latency. Batch process embeddings to cut API calls. Cache frequent queries and responses. Fine-tune with domain-specific data if available. Monitor latency and cost via Vertex AI’s tools, and use asynchronous calls for non-blocking operations.

HuggingFace all-mpnet-base-v2 optimization tips

To optimize the all-mpnet-base-v2 model in a RAG setup, preprocess input text by removing noise, truncating to the 384-token limit, and splitting documents into contextually coherent chunks. Use cosine similarity for retrieval, as the model is fine-tuned for this metric. Batch embedding generation improves throughput, while leveraging GPU acceleration reduces latency. Fine-tune the model on domain-specific data for better relevance. Index embeddings with FAISS or HNSW for efficient nearest-neighbor searches, and experiment with chunk overlap to balance context retention and redundancy.

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 power of building a RAG system from the ground up—and that’s no small feat! In this tutorial, you learned how to weave together Haystack as your orchestration framework, Milvus as your lightning-fast vector database, Google Vertex AI Gemini 2.0 Flash as your efficient large language model (LLM), and HuggingFace’s all-mpnet-base-v2 as your embedding model to create a seamless retrieval-augmented generation pipeline. You saw how Haystack acts as the glue, connecting data pipelines and streamlining workflows, while Milvus handles the heavy lifting of storing and retrieving vector embeddings at scale. The all-mpnet-base-v2 model transformed your raw text into rich numerical representations, enabling semantic search, and Gemini 2.0 Flash stepped in to generate human-like responses grounded in the retrieved context. Together, these tools turned abstract concepts into a functional system that can answer questions, summarize content, or even power chatbots with precision and speed!

But you didn’t stop there—you also picked up pro tips for optimizing your RAG pipeline, like tuning retrieval parameters and balancing cost-performance tradeoffs. The free RAG cost calculator shared in the tutorial gave you a practical way to estimate expenses and scale your projects wisely. Now, imagine what’s next: customizing this foundation for your own datasets, experimenting with different LLMs, or integrating domain-specific embeddings. The skills you’ve gained aren’t just about code—they’re about solving real-world problems creatively. So go ahead, dive back in, tweak those knobs, and watch your ideas come to life. The world of intelligent applications is yours to shape, and you’ve got everything you need to build something amazing. Let’s get coding—your next breakthrough is just a RAG pipeline away! 🚀

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! 💖