server

app.py

@@ -3,8 +3,9 @@ import json
 import requests
 import sseclient
 import gradio as gr
+import server
 
-API_URL = "http://localhost:8000/v1/completions"
+API_URL = "http://localhost:8000/v1/chat/completions"
 
 
 def stream_completion(message, history, max_tokens, temperature, top_p, beta):

server.py

@@ -33,4 +33,4 @@ setup_mixinputs()
 threading.Thread(target=launch_vllm_server, daemon=True).start()
 
 # Step 3: Give time for server to initialize
-time.sleep(20)
+time.sleep(60)