Update inference.py

inference.py

@@ -6,6 +6,8 @@ from evo_model import EvoTransformerV22
 from search_utils import web_search
 import openai
 import time
+import psutil
+import platform
 
 # 🔐 Load OpenAI API Key securely
 openai.api_key = os.getenv("OPENAI_API_KEY")
@@ -92,3 +94,18 @@ def get_model_config():
         "ffn_dim": 1024,
         "memory_enabled": True
     }
+
+def get_system_stats():
+    gpu_info = torch.cuda.get_device_properties(0) if torch.cuda.is_available() else None
+    memory = psutil.virtual_memory()
+
+    return {
+        "device": "GPU" if torch.cuda.is_available() else "CPU",
+        "cpu_usage_percent": psutil.cpu_percent(),
+        "memory_used_gb": round(memory.used / (1024 ** 3), 2),
+        "memory_total_gb": round(memory.total / (1024 ** 3), 2),
+        "gpu_name": gpu_info.name if gpu_info else "N/A",
+        "gpu_memory_total_gb": round(gpu_info.total_memory / (1024 ** 3), 2) if gpu_info else "N/A",
+        "gpu_memory_used_gb": round(torch.cuda.memory_allocated() / (1024 ** 3), 2) if gpu_info else "N/A",
+        "platform": platform.platform()
+    }