Multimodal RAG Pipeline

from mixpeek import Mixpeek
import openai

client = Mixpeek(api_key="YOUR_API_KEY")

# 1. Build knowledge base
namespace = client.namespaces.create(name="rag-kb")
collection = client.collections.create(
    namespace_id=namespace.id,
    name="docs-and-media",
    extractors=["text-embedding-v2", "image-embedding-v2"],
    chunk_strategy="semantic"
)

# 2. Ingest your content
client.buckets.upload(
    collection_id=collection.id,
    url="s3://your-bucket/knowledge-base/"
)

# 3. Retrieve + Generate
def rag_query(question: str):
    # Retrieve relevant context
    results = client.retrievers.execute(
        retriever_id=retriever.id,
        query=question,
        settings={"limit": 5}
    )

    # Build context from results
    context = "\n".join([r.content for r in results])

    # Generate answer with LLM
    response = openai.chat.completions.create(
        model="gpt-4",
        messages=[
            {"role": "system", "content": f"Answer based on this context:\n{context}"},
            {"role": "user", "content": question}
        ]
    )
    return response.choices[0].message.content