diff --git a/getting-started/build_with_llama_4.ipynb b/getting-started/build_with_llama_4.ipynb
index 30d7aaa9f..231820bda 100644
--- a/getting-started/build_with_llama_4.ipynb
+++ b/getting-started/build_with_llama_4.ipynb
@@ -52,7 +52,7 @@
     "* Maverick which has 17B x 128 Experts MoE \n",
     "\n",
     "Long context window : \n",
-    "-- If you want to used this model on Single GPU with 10M context =  Need to use an INT4-quantized version of Llama 4 Scout on 1xH100 GPU \n",
+    "-- If you want to use this model on Single GPU with 10M context =  Need to use an INT4-quantized version of Llama 4 Scout on 1xH100 GPU \n",
     "\n",
     "\n",
     "Hardware requirements without quantized models :\n",