Add GPU debug logging

app.py

@@ -17,9 +17,24 @@ try:
     from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
     from peft import PeftModel
     GPU_AVAILABLE = torch.cuda.is_available()
-except ImportError:
+
+    # Debug info
+    print("=" * 50)
+    print(f"PyTorch version: {torch.__version__}")
+    print(f"CUDA available: {torch.cuda.is_available()}")
+    if torch.cuda.is_available():
+        print(f"CUDA version: {torch.version.cuda}")
+        print(f"GPU count: {torch.cuda.device_count()}")
+        print(f"GPU name: {torch.cuda.get_device_name(0)}")
+    else:
+        print("CUDA not available - checking why...")
+        print(f"torch.backends.cudnn.enabled: {torch.backends.cudnn.enabled if hasattr(torch.backends, 'cudnn') else 'N/A'}")
+    print("=" * 50)
+
+except ImportError as e:
     GPU_AVAILABLE = False
-    print("Warning: torch/transformers not available, running in mock mode")
+    print(f"Warning: Import error - {e}")
+    print("Running in mock mode")
 
 # ============================================================
 # 모델 레지스트리 (HF Hub 경로)