diff --git a/src/model.rs b/src/model.rs
index 2f4d28d..dc17fff 100644
--- a/src/model.rs
+++ b/src/model.rs
@@ -1,3 +1,4 @@
+use ort::execution_providers::CUDAExecutionProvider;
 use ort::session::{Session, builder::GraphOptimizationLevel};
 use sha2::{Digest, Sha256};
 use std::error::Error;
@@ -156,16 +157,17 @@ pub fn get_model_path(
     Ok(model_path)
 }
 
-/// Create an ONNX Runtime session from model path
+/// Create an ONNX Runtime session from model path with CUDA backend
 pub fn create_session(model_path: &Path) -> Result<Session, Box<dyn Error>> {
-    println!("Loading model into ONNX Runtime...");
+    println!("Loading model into ONNX Runtime with CUDA backend...");
 
     let session = Session::builder()?
+        .with_execution_providers([CUDAExecutionProvider::default().build()])?
         .with_optimization_level(GraphOptimizationLevel::Level3)?
         .with_intra_threads(4)?
         .commit_from_file(model_path)?;
 
-    println!("Model loaded successfully!");
+    println!("Model loaded successfully with CUDA!");
 
     Ok(session)
 }