diff --git a/unsloth/models/llama.py b/unsloth/models/llama.py
index 8024c915ff..7211015218 100644
--- a/unsloth/models/llama.py
+++ b/unsloth/models/llama.py
@@ -680,7 +680,6 @@ pass
 
 # https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/modeling_llama.py#L825
 @torch.compile
-@torch.inference_mode
 def LlamaModel_fast_forward_inference(
     self,
     input_ids,