Update app.py

app.py

@@ -1,28 +1,21 @@
-import httpx
-from llama_index.llms.ollama import Ollama
-from llama_index.core.llms import ChatMessage
-import gradio as gr
-
 import os
 import threading
 import subprocess
+import gradio as gr
+from llama_index.llms.ollama import Ollama
+from llama_index.core.llms import ChatMessage
 
-def ollama():
+# Fonction pour démarrer le serveur Ollama
+def start_ollama():
     os.environ['OLLAMA_HOST'] = '127.0.0.1:11434'
     os.environ['OLLAMA_ORIGINS'] = '*'
-    subprocess.Popen(["ollama", "serve"])
+    subprocess.run(["ollama", "serve"])
 
-ollama_thread = threading.Thread(target=ollama)
+# Démarrer le serveur Ollama dans un thread séparé
+ollama_thread = threading.Thread(target=start_ollama)
 ollama_thread.start()
 
-proxy_url = "http://localhost:11434"
-
-
-proxies = {
-    "http://": "http://localhost:11434",
-    "https://": "http://localhost:11434",
-}
-
+# Initialiser le modèle Ollama
 llm = Ollama(model="llama3", request_timeout=120.0)
 
 def get_completion(prompt):
@@ -56,4 +49,5 @@ chat_interface = gr.Interface(fn=generate_chat_response, inputs=chat_input, outp
 
 app = gr.TabbedInterface([single_interface, chat_interface], ["Single Completion", "Chat"])
 
-app.launch()
+if __name__ == "__main__":
+    app.launch(server_name="0.0.0.0", server_port=7860)