quick fix to use openai instead of ollama, in vetor_storage.py

services/rag/langchain/QWEN.md

@@ -105,12 +105,19 @@ The project is organized into 8 development phases as outlined in `PLANNING.md`:
 The project uses environment variables for configuration:
 
 ```env
+# Embedding configuration
 OLLAMA_EMBEDDING_MODEL=MODEL      # Name of the Ollama model for embeddings
+OPENAI_EMBEDDING_MODEL=MODEL      # Name of the OpenAI model for embeddings (default: text-embedding-ada-002)
+OPENAI_EMBEDDING_BASE_URL=URL     # OpenAI-compatible API URL for embeddings
+OPENAI_EMBEDDING_API_KEY=KEY      # API key for OpenAI-compatible embedding service
+EMBEDDING_STRATEGY=ollama         # Strategy to use for embeddings: "ollama" (default) or "openai"
+
+# Chat model configuration
 OLLAMA_CHAT_MODEL=MODEL           # Name of the Ollama model for chat
-OPENAI_CHAT_URL=URL               # OpenAI-compatible API URL
-OPENAI_CHAT_KEY=KEY               # Authorization token for OpenAI-compatible API
-OPENAI_CHAT_MODEL=MODEL           # Name of the OpenAI-compatible model to use
-CHAT_MODEL_STRATEGY=ollama        # Strategy to use: "ollama" (default) or "openai"
+OPENAI_CHAT_URL=URL               # OpenAI-compatible API URL for chat
+OPENAI_CHAT_KEY=KEY               # Authorization token for OpenAI-compatible API for chat
+OPENAI_CHAT_MODEL=MODEL           # Name of the OpenAI-compatible model to use for chat
+CHAT_MODEL_STRATEGY=ollama        # Strategy to use for chat: "ollama" (default) or "openai"
 ```
 
 ## Building and Running
@@ -235,3 +242,5 @@ The project is in early development phase. The virtual environment is set up and
 - If encountering "No module named 'unstructured_inference'" error, install unstructured-inference
 - If seeing OCR-related errors, ensure tesseract is installed at the system level and unstructured-pytesseract is available
 - For language detection issues, verify that appropriate spaCy models are downloaded
+- If getting Ollama connection errors when using OpenAI strategy, ensure EMBEDDING_STRATEGY is set correctly in .env
+- When deploying without Ollama, set both CHAT_MODEL_STRATEGY and EMBEDDING_STRATEGY to "openai" in your .env file

services/rag/langchain/vector_storage.py

@@ -1,4 +1,4 @@
-"""Vector storage module using Qdrant and Ollama embeddings for the RAG solution."""
+"""Vector storage module using Qdrant and configurable embeddings for the RAG solution."""
 
 import os
 from typing import Optional
@@ -7,6 +7,7 @@ from dotenv import load_dotenv
 from langchain_qdrant import QdrantVectorStore
 from langchain_core.documents import Document
 from langchain_ollama import OllamaEmbeddings
+from langchain_openai import OpenAIEmbeddings
 from qdrant_client import QdrantClient
 
 # Load environment variables
@@ -17,15 +18,19 @@ QDRANT_HOST = os.getenv("QDRANT_HOST", "localhost")
 QDRANT_REST_PORT = int(os.getenv("QDRANT_REST_PORT", 6333))
 QDRANT_GRPC_PORT = int(os.getenv("QDRANT_GRPC_PORT", 6334))
 
-# Ollama embedding model configuration
+# Embedding model configuration
+EMBEDDING_STRATEGY = os.getenv("EMBEDDING_STRATEGY", "ollama").lower()
 OLLAMA_EMBEDDING_MODEL = os.getenv("OLLAMA_EMBEDDING_MODEL", "nomic-embed-text")
+OPENAI_EMBEDDING_MODEL = os.getenv("OPENAI_EMBEDDING_MODEL", "text-embedding-ada-002")
+OPENAI_EMBEDDING_BASE_URL = os.getenv("OPENAI_EMBEDDING_BASE_URL")
+OPENAI_EMBEDDING_API_KEY = os.getenv("OPENAI_EMBEDDING_API_KEY")
 
 
 def initialize_vector_store(
     collection_name: str = "documents_langchain", recreate_collection: bool = False
 ) -> QdrantVectorStore:
     """
-    Initialize and return a Qdrant vector store with Ollama embeddings.
+    Initialize and return a Qdrant vector store with configurable embeddings.
 
     Args:
         collection_name: Name of the Qdrant collection to use
@@ -34,11 +39,24 @@ def initialize_vector_store(
     Returns:
         Initialized Qdrant vector store
     """
-    # Initialize Ollama embeddings
-    embeddings = OllamaEmbeddings(
-        model=OLLAMA_EMBEDDING_MODEL,
-        base_url="http://localhost:11434",  # Default Ollama URL
-    )
+    # Determine which embedding strategy to use
+    if EMBEDDING_STRATEGY == "openai":
+        # Validate required OpenAI embedding variables
+        if not OPENAI_EMBEDDING_API_KEY or not OPENAI_EMBEDDING_BASE_URL:
+            raise ValueError("OPENAI_EMBEDDING_API_KEY and OPENAI_EMBEDDING_BASE_URL must be set when using OpenAI embedding strategy")
+
+        # Initialize OpenAI embeddings
+        embeddings = OpenAIEmbeddings(
+            model=OPENAI_EMBEDDING_MODEL,
+            openai_api_base=OPENAI_EMBEDDING_BASE_URL,
+            openai_api_key=OPENAI_EMBEDDING_API_KEY,
+        )
+    else:  # Default to ollama
+        # Initialize Ollama embeddings
+        embeddings = OllamaEmbeddings(
+            model=OLLAMA_EMBEDDING_MODEL,
+            base_url="http://localhost:11434",  # Default Ollama URL
+        )
 
     # Check if collection exists and create if needed
     client = QdrantClient(