Update server.py

server.py

@@ -5,11 +5,45 @@ from llama_index.core import (
     StorageContext,
     load_index_from_storage,
 )
+from llama_index.core import Settings
+from llama_index.llms.azure_openai import AzureOpenAI
+from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding
 from typing import Dict, Optional
 import json
 import os
 import aiohttp  # Necesario para las peticiones HTTP asíncronas
 
+
+
+
+api_key = os.environ.get('AZURE_API_KEY')
+azure_endpoint = "https://pharmaia-gpt.openai.azure.com/"
+api_version = "2024-02-01"
+
+llm = AzureOpenAI(
+    model="gpt-4.1",
+    deployment_name="gpt-4.1",
+    api_key=api_key,
+    azure_endpoint=azure_endpoint,
+    api_version=api_version,
+)
+# You need to deploy your own embedding model as well as your own chat completion model
+embed_model = AzureOpenAIEmbedding(
+    model="text-embedding-3-large",
+    deployment_name="text-embedding-3-large",
+    api_key=api_key,
+    azure_endpoint=azure_endpoint,
+    api_version=api_version,
+)
+
+Settings.llm = llm
+Settings.embed_model = embed_model
+
+
+
+
+
+
 port = int(os.getenv("PORT", 7860))
 mcp = FastMCP("OnBase", port=port)
 
@@ -56,8 +90,7 @@ for source, rel_path in SOURCES.items():
     uri="retriever://documentos/{fuente}",
     name="DocumentRetriever",
     description="Retrieve documents from different regulations using semantic search.",
-    mime_type="application/json",
-    tags={"llm", "retrieval"}
+    mime_type="application/json"
 )
 def retrieve_docs(query: str, fuente: str = 'oms',  top_k: int = 3) -> dict:
     """