modified to combine openwebui and ollama

README.md

@@ -74,8 +74,8 @@ To deploy the service:
 ## API Documentation
 
 Once the service is running, you can access the API documentation at:
-- Swagger UI: http://localhost:5251/docs
+- Swagger UI: http://localhost:5252/docs
-- ReDoc: http://localhost:5251/redoc
+- ReDoc: http://localhost:5252/redoc
 
 ## OpenWebUI Configuration
 
@@ -83,8 +83,8 @@ To configure OpenWebUI to use this service as its backend:
 
 1. Set the following environment variables in OpenWebUI:
    ```
-   OLLAMA_API_BASE_URL=http://your-server-ip:5251/ollama
+   OLLAMA_API_BASE_URL=http://your-server-ip:5252/ollama
-   OPENAI_API_BASE_URL=http://your-server-ip:5251/api
+   OPENAI_API_BASE_URL=http://your-server-ip:5252/api
    ```
 
 2. Restart OpenWebUI to apply the changes.

ai_service/README.md

@@ -52,8 +52,8 @@ ai_service/
 ## API Documentation
 
 Once the service is running, you can access the API documentation at:
-- Swagger UI: http://localhost:5251/docs
+- Swagger UI: http://localhost:5252/docs
-- ReDoc: http://localhost:5251/redoc
+- ReDoc: http://localhost:5252/redoc
 
 ## Deployment
 
@@ -69,7 +69,7 @@ To deploy the service:
    ./deploy.sh
    ```
 
-This will start the service on port 5251 using uvicorn with nohup.
+This will start the service on port 5252 using uvicorn with nohup.
 
 ## API Endpoints
 

ai_service/config.py

@@ -6,14 +6,16 @@ import os
 from dotenv import load_dotenv
 
 # Load environment variables from .env file
-load_dotenv()
+import os.path
+dotenv_path = os.path.join(os.path.dirname(__file__), '.env')
+load_dotenv(dotenv_path=dotenv_path)
 
 class Config:
     """Base configuration."""
 
     # API configuration
     API_HOST = os.environ.get('API_HOST', '0.0.0.0')
-    API_PORT = int(os.environ.get('API_PORT', 5251))
+    API_PORT = int(os.environ.get('API_PORT', 5252))
 
     # OpenWebUI configuration
     OPENWEBUI_URL = os.environ.get('OPENWEBUI_URL', 'http://104.225.217.215:8080')

ai_service/data/chats.json

@@ -76,5 +76,127 @@
       }
     ],
     "team_members": []
+  },
+  "b72d5056-4061-40e5-9d8f-11c479ff95ed": {
+    "id": "b72d5056-4061-40e5-9d8f-11c479ff95ed",
+    "title": "Test Chat",
+    "user_id": "test_user",
+    "model_id": "llama3.1",
+    "is_team_chat": false,
+    "created_at": "2025-05-09T16:52:09.064723",
+    "updated_at": "2025-05-09T16:52:14.572522",
+    "messages": [
+      {
+        "id": "7a425092-90a1-447f-b424-f92c001dd70a",
+        "content": "Hello, how are you?",
+        "user_id": "test_user",
+        "is_user_message": true,
+        "timestamp": "2025-05-09T16:52:14.548916"
+      },
+      {
+        "id": "00927dda-01f8-4f1a-808e-1b596dca258e",
+        "content": "Error generating response: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/generate (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x11101fed0>: Failed to establish a new connection: [Errno 61] Connection refused'))",
+        "user_id": null,
+        "is_user_message": false,
+        "timestamp": "2025-05-09T16:52:14.572493"
+      }
+    ],
+    "team_members": []
+  },
+  "36aed910-1c6e-4184-84f2-729227d57915": {
+    "id": "36aed910-1c6e-4184-84f2-729227d57915",
+    "title": "Test Chat",
+    "user_id": "test_user",
+    "model_id": "llama3.1",
+    "is_team_chat": false,
+    "created_at": "2025-05-09T18:32:34.034219",
+    "updated_at": "2025-05-09T18:40:42.597983",
+    "messages": [
+      {
+        "id": "cbb1e7d2-6b66-4d0a-9907-0b72d344f3f3",
+        "content": "Hello, how are you?",
+        "user_id": "test_user",
+        "is_user_message": true,
+        "timestamp": "2025-05-09T18:32:39.388163"
+      },
+      {
+        "id": "6b2abf18-f937-4390-bf51-ef46115f2291",
+        "content": "Error generating response: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/generate (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x11154f510>: Failed to establish a new connection: [Errno 61] Connection refused'))",
+        "user_id": null,
+        "is_user_message": false,
+        "timestamp": "2025-05-09T18:32:39.408236"
+      },
+      {
+        "id": "71c388c9-1e96-4b18-a6ac-5f69082e3fa0",
+        "content": "Hello, how are you?",
+        "user_id": "test_user",
+        "is_user_message": true,
+        "timestamp": "2025-05-09T18:33:48.306212"
+      },
+      {
+        "id": "b2cfa76f-c4bc-4d02-89b5-ebcf0a24616a",
+        "content": "Error generating response: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/generate (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x1068eb610>: Failed to establish a new connection: [Errno 61] Connection refused'))",
+        "user_id": null,
+        "is_user_message": false,
+        "timestamp": "2025-05-09T18:33:48.321634"
+      },
+      {
+        "id": "4e61326b-d457-4087-a0e9-a8416de42a51",
+        "content": "Hello, how are you?",
+        "user_id": "test_user",
+        "is_user_message": true,
+        "timestamp": "2025-05-09T18:36:15.153049"
+      },
+      {
+        "id": "f63c7ee3-9ade-4543-88be-129a48213c85",
+        "content": "Error generating response: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/generate (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x10ac36ad0>: Failed to establish a new connection: [Errno 61] Connection refused'))",
+        "user_id": null,
+        "is_user_message": false,
+        "timestamp": "2025-05-09T18:36:15.174809"
+      },
+      {
+        "id": "d463b173-8a9f-4972-9293-380f2959f2e3",
+        "content": "Hello, how are you?",
+        "user_id": "test_user",
+        "is_user_message": true,
+        "timestamp": "2025-05-09T18:38:17.218487"
+      },
+      {
+        "id": "ee7c8a2e-c7fe-4545-b9d0-895f33fa2f62",
+        "content": "Error generating response: HTTPConnectionPool(host='104.225.217.215', port=11434): Max retries exceeded with url: /api/generate (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x10e343410>: Failed to establish a new connection: [Errno 61] Connection refused'))",
+        "user_id": null,
+        "is_user_message": false,
+        "timestamp": "2025-05-09T18:38:17.493977"
+      },
+      {
+        "id": "3b94da97-e224-41c5-95c1-6d17118e5665",
+        "content": "Hello, how are you?",
+        "user_id": "test_user",
+        "is_user_message": true,
+        "timestamp": "2025-05-09T18:39:29.063581"
+      },
+      {
+        "id": "2843b17c-35bb-49cd-b306-cc553712341c",
+        "content": "Error generating response: 401 Client Error: Unauthorized for url: http://104.225.217.215:8080/ollama/api/generate",
+        "user_id": null,
+        "is_user_message": false,
+        "timestamp": "2025-05-09T18:39:29.605996"
+      },
+      {
+        "id": "314a855a-9922-4059-bb02-57f56baa8c62",
+        "content": "Hello, how are you?",
+        "user_id": "test_user",
+        "is_user_message": true,
+        "timestamp": "2025-05-09T18:40:42.061269"
+      },
+      {
+        "id": "e417708d-04b6-4b50-999d-cdc242a2283b",
+        "content": "Error generating response: 401 Client Error: Unauthorized for url: http://104.225.217.215:8080/ollama/api/generate",
+        "user_id": null,
+        "is_user_message": false,
+        "timestamp": "2025-05-09T18:40:42.597967"
+      }
+    ],
+    "team_members": []
   }
 }

ai_service/deploy.sh

@@ -8,7 +8,7 @@ source venv/bin/activate
 
 # Export environment variables
 export API_HOST=0.0.0.0
-export API_PORT=5251
+export API_PORT=5252
 
 # Make sure the Python path includes the current directory
 export PYTHONPATH=$PYTHONPATH:$(pwd)

ai_service/models/model_service.py

@@ -2,13 +2,10 @@
 Service for model management and interaction.
 """
 
-import os
-import json
 import requests
 from typing import List, Dict, Any, Optional
 
 from ai_service.config import config
-from ai_service.embeddings.document_service import document_service
 from ai_service.models.model_parameters import ModelParameters
 
 class ModelService:
@@ -112,8 +109,8 @@ class ModelService:
         if model_id not in self.AVAILABLE_MODELS:
             model_id = self.default_model
 
-        # Get the provider for this model
+        # Ensure we're using a valid model
-        provider = self.AVAILABLE_MODELS[model_id].get('provider', 'ollama')
+        # (model_id is already validated above)
 
         # Prepare the messages for the API call
         messages = []
@@ -212,15 +209,13 @@ class ModelService:
 
         # Make the API call to Ollama
         try:
-            # Prepare headers with API key
+            # Prepare headers
             headers = {"Content-Type": "application/json"}
-            if self.openwebui_api_key:
-                headers["Authorization"] = f"Bearer {self.openwebui_api_key}"
 
-            # Ollama API endpoint is /api/chat or /api/generate
+            # Direct Ollama API call
             response = requests.post(
-                f"{self.ollama_api_url}/api/generate",
+                f"{self.ollama_api_url}/api/chat",
-                headers=headers,
+                headers={"Content-Type": "application/json"},
                 json=request_json,
                 timeout=60
             )
@@ -229,18 +224,13 @@ class ModelService:
             result = response.json()
 
             # Extract the response content from Ollama
-            # The response format depends on whether we're using /api/chat or /api/generate
+            # The response format for Ollama API
             if 'message' in result and 'content' in result['message']:
-                # Format for /api/chat
                 return result['message']['content']
-            elif 'response' in result:
-                # Format for /api/generate
-                return result['response']
             else:
                 return "Error: Unexpected response format from Ollama"
 
         except Exception as e:
-            print(f"Error calling Ollama API: {str(e)}")
             return f"Error generating response: {str(e)}"
 
 

deploy_ai_service.sh

@@ -12,7 +12,7 @@ touch ai_service/data/document_metadata.json
 touch ai_service/data/chats.json
 
 # Start the AI service
-echo "Starting AI Service on port 5251..."
+echo "Starting AI Service on port 5252..."
 VENV_PATH="./venv"
 PYTHON_PATH="$VENV_PATH/bin/python"
 
@@ -30,13 +30,13 @@ sleep 2
 
 # Check if the service is running
 if pgrep -f "run_ai_service.py" > /dev/null; then
-    echo "AI Service started successfully on port 5251"
+    echo "AI Service started successfully on port 5252"
     echo "Check ai_service.log for output"
     echo "To stop the service, run: pkill -f \"run_ai_service.py\""
 
     # Test the health endpoint
     echo -e "\nTesting health endpoint..."
-    curl -s http://localhost:5251/health
+    curl -s http://localhost:5252/health
     echo -e "\n"
 
     # Test creating a chat and sending a message
@@ -49,7 +49,7 @@ import json
 
 # Create a chat
 response = requests.post(
-    "http://localhost:5251/chats",
+    "http://localhost:5252/chats",
     json={
         "user_id": "test_user",
         "title": "Test Chat",
@@ -63,7 +63,7 @@ if response.status_code == 200:
 
     # Send a message with parameters
     response = requests.post(
-        f"http://localhost:5251/chats/{chat_id}/messages",
+        f"http://localhost:5252/chats/{chat_id}/messages",
         json={
             "message": "Hello, AI!",
             "user_id": "test_user",

remote_deploy.sh

@@ -64,7 +64,7 @@ sleep 5
 ssh -p $SERVER_PORT $SERVER_USER@$SERVER_IP "ps aux | grep 'uvicorn ai_service.api:app' | grep -v grep"
 if [ $? -eq 0 ]; then
     echo "Service is running!"
-    echo "You can access the API at: http://$SERVER_IP:5251"
+    echo "You can access the API at: http://$SERVER_IP:5252"
     echo "Check logs with: ssh -p $SERVER_PORT $SERVER_USER@$SERVER_IP 'tail -f $REMOTE_DIR/ai_service_app/logs/ai_service.log'"
 else
     echo "Error: Service failed to start. Check logs on the server."

run_ai_service.py

@@ -8,10 +8,10 @@ import sys
 
 # Run the service
 if __name__ == "__main__":
-    print("Starting AI service on 0.0.0.0:5251")
+    print("Starting AI service on 0.0.0.0:5252")
     uvicorn.run(
         "ai_service.api:app",
         host="0.0.0.0",
-        port=5251,
+        port=5252,
         reload=True
     )