Using Haystack

Haystack provides a robust framework for implementing Retrieval-Augmented Generation (RAG) by integrating data retrieval and response generation components. This section details how to set up and use Haystack for efficient data retrieval and response generation with MonsterAPI.

Follow these steps to use our LLM endpoints with Haystack:

Step 1: Setting Up the Components

1. LinkContentFetcher: Fetches HTML content from URLs.

2. HTMLToDocument: Converts HTML content into document objects.

3. PromptBuilder: Prepares a prompt using the content and user’s query.

4. OpenAIGenerator: Generates responses using MonsterAPI.

   Python

   from haystack import Pipeline
   from haystack.utils import Secret
   from haystack.components.fetchers import LinkContentFetcher
   from haystack.components.converters import HTMLToDocument
   from haystack.components.builders import PromptBuilder
   from haystack.components.generators import OpenAIGenerator
   
   fetcher = LinkContentFetcher()
   converter = HTMLToDocument()
   
   prompt_template = """
   According to the contents of this website:
   {% for document in documents %}
    {{document.content}}
   {% endfor %}
   Answer the given question: {{query}}
   Answer:
   """
   
   prompt_builder = PromptBuilder(template=prompt_template)
   
   llm = OpenAIGenerator(
       api_key=Secret.from_env_var("MONSTER_API_KEY"),
       api_base_url="https://llm.monsterapi.ai/v1/",
       model="microsoft/Phi-3-mini-4k-instruct",
       generation_kwargs = {"max_tokens": 256}
   )
   

Step 2: Building and Connecting the Pipeline

1. Create the Pipeline: Add components sequentially.

2. Connect Components: Ensure smooth data flow between components.

   Python

   pipeline = Pipeline()
   pipeline.add_component("fetcher", fetcher)
   pipeline.add_component("converter", converter)
   pipeline.add_component("prompt", prompt_builder)
   pipeline.add_component("llm", llm)
   

Step 3: Running the Pipeline

1. Execute the Pipeline: Run the pipeline with necessary inputs and retrieve the response.

   Python

   result = pipeline.run({
       "fetcher": {"urls": ["https://developer.monsterapi.ai/docs/"]},
       "prompt": {"query": "What are the features of MonsterAPI?"}
   })
   
   print(result["llm"]["replies"][0])
   

By following these steps, you can effectively utilize Haystack with our platform to create a RAG system that dynamically retrieves relevant information and generates accurate, context-aware responses.