x

scripts/start_vllm.py

@@ -325,7 +325,21 @@ def configure_and_launch(model_idx, gpu_count):
     print(f" Command:   {' '.join(cmd)}")
     print("="*60 + "\n")
     
-    os.execvpe("vllm", cmd, env)
+    # Check if model path exists
+    if not os.path.exists(model_path):
+        print(f"ERROR: Model path does not exist: {model_path}")
+        print(f"Please ensure the model is mounted at {model_path}")
+        sys.exit(1)
+    
+    # Run vllm serve
+    try:
+        result = subprocess.run(cmd, env=env, check=False)
+        if result.returncode != 0:
+            print(f"\nERROR: vllm serve exited with code {result.returncode}")
+            sys.exit(result.returncode)
+    except Exception as e:
+        print(f"\nERROR: Failed to start vllm serve: {e}")
+        sys.exit(1)
 
 def main():
     check_dependencies()