Multimodal Knowledge Base

from mixpeek import Mixpeek
from openai import OpenAI

client = Mixpeek(api_key="YOUR_API_KEY")
openai = OpenAI(api_key="YOUR_OPENAI_KEY")

# Create collections for each content type
docs_col = client.collections.create(
    namespace_id="ns_your_namespace",
    name="knowledge_docs",
    extractors=["document-graph-extractor", "text-extractor"]
)

videos_col = client.collections.create(
    namespace_id="ns_your_namespace",
    name="knowledge_videos",
    extractors=["multimodal-extractor", "text-extractor"]
)

# Build unified retriever spanning all collections
retriever = client.retrievers.create(
    namespace_id="ns_your_namespace",
    name="knowledge_base",
    collection_ids=["col_knowledge_docs", "col_knowledge_videos"],
    stages=[
        {"type": "feature_search", "top_k": 50},
        {"type": "rerank", "top_k": 10},
        {"type": "rag_prepare"}
    ]
)

# Ask a question across all content
results = client.retrievers.execute(
    retriever_id=retriever["retriever_id"],
    query={"text": "What is our company policy on remote work?"}
)

context = "\n".join([
    f"[{i+1}] {doc['text']} (Source: {doc['root_object_id']}, Type: {doc['modality']})"
    for i, doc in enumerate(results["documents"])
])

response = openai.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": f"Answer from this knowledge base:\n{context}"},
        {"role": "user", "content": "What is our company policy on remote work?"}
    ]
)
print(response.choices[0].message.content)