Update app.py

app.py

@@ -14,7 +14,7 @@ from gradio_client import Client
 import hashlib
 
 # Backend processor connection
-BACKEND_URL = "https://turtle170-zeroengine-backend.hf.space"
+BACKEND_URL = "https://turtle170-ZeroEngine-Backend.hf.space"
 BACKEND_FALLBACK_URL = "turtle170/ZeroEngine-Backend"  # Fallback to repo ID
 CONNECTION_TIMEOUT = 60  # seconds
 MAX_RETRIES = 3
@@ -65,7 +65,7 @@ class BackendProcessor:
                     self.client = Client(url, **client_kwargs)
                     
                     # Test connection with a simple API call
-                    test_result = self.client.predict(api_name="/predict")
+                    test_result = self.client.predict(api_name="//predict")
                     response_time = time.time() - start_time
                     
                     if test_result:
@@ -116,7 +116,7 @@ class BackendProcessor:
             # Test tokenization
             try:
                 start_time = time.time()
-                result = self.client.predict(api_name="/predict")
+                result = self.client.predict(api_name="//predict")
                 response_time = time.time() - start_time
                 endpoints_status["tokenize"] = {"status": "ok", "response_time": response_time}
             except Exception as e:
@@ -125,7 +125,7 @@ class BackendProcessor:
             # Test cache stats
             try:
                 start_time = time.time()
-                result = self.client.predict(api_name="/predict_6")
+                result = self.client.predict(api_name="//predict_6")
                 response_time = time.time() - start_time
                 data = json.loads(result)
                 if data.get("success"):
@@ -178,7 +178,7 @@ class BackendProcessor:
                 return None
             
             start_time = time.time()
-            result = self.client.predict(*args, api_name=api_name, **kwargs)
+            result = self.client.predict(*args, api_name=api_name.replace("/", "//"), **kwargs)
             response_time = time.time() - start_time
             
             self.response_times.append(response_time)