Add dedicated page for running Ollama models

app/api/models.py

@@ -119,6 +119,54 @@ async def get_models():
             detail="Errore nel recupero dei modelli"
         )
 
+
+@router.get("/models/running")
+async def get_running_models() -> Dict[str, Any]:
+    """
+    Recupera i modelli attualmente residenti in memoria, equivalenti a `ollama ps`.
+
+    Returns:
+        Dict[str, Any]: Payload con modelli running e conteggio
+    """
+    try:
+        response = requests.get(
+            f"{settings.OLLAMA_HOST}/api/ps",
+            timeout=settings.OLLAMA_TIMEOUT
+        )
+
+        if response.status_code != 200:
+            raise HTTPException(
+                status_code=502,
+                detail="Ollama non disponibile"
+            )
+
+        data = response.json()
+        models_data = data.get("models", [])
+
+        return {
+            "models": models_data,
+            "total": len(models_data)
+        }
+
+    except requests.exceptions.Timeout:
+        raise HTTPException(
+            status_code=504,
+            detail="Timeout: Ollama non ha risposto in tempo"
+        )
+    except requests.exceptions.ConnectionError:
+        raise HTTPException(
+            status_code=502,
+            detail="Impossible connettersi a Ollama"
+        )
+    except HTTPException:
+        raise
+    except Exception as e:
+        logger.error(f"Error fetching running models: {e}")
+        raise HTTPException(
+            status_code=500,
+            detail="Errore nel recupero dei modelli residenti"
+        )
+
 @router.get("/models/{model_name}", response_model=ModelInfo)
 async def get_model(model_name: str):
     """