diff --git a/wgpu/examples/hello-compute/main.rs b/wgpu/examples/hello-compute/main.rs
index 46d6d70ba1..0ddf645831 100644
--- a/wgpu/examples/hello-compute/main.rs
+++ b/wgpu/examples/hello-compute/main.rs
@@ -22,8 +22,8 @@ async fn execute_gpu(numbers: Vec<u32>) -> Vec<u32> {
     let slice_size = numbers.len() * std::mem::size_of::<u32>();
     let size = slice_size as wgpu::BufferAddress;
 
-    let instace = wgpu::Instance::new();
-    let adapter = instace
+    let instance = wgpu::Instance::new();
+    let adapter = instance
         .request_adapter(
             &wgpu::RequestAdapterOptions {
                 power_preference: wgpu::PowerPreference::Default,
@@ -154,6 +154,12 @@ fn main() {
 mod tests {
     use super::*;
 
+    #[test]
+    fn test_compute_0() {
+        let input = vec![];
+        futures::executor::block_on(assert_execute_gpu(input, vec![]));
+    }
+
     #[test]
     fn test_compute_1() {
         let input = vec![1, 2, 3, 4];