From dcc83ea841ab6100d6b47a070329e1ba4cf78752 Mon Sep 17 00:00:00 2001
From: Omar Sanseviero <osanseviero@users.noreply.huggingface.co>
Date: Fri, 4 Apr 2025 13:12:40 +0000
Subject: [PATCH] Update README.md (#13)

- Update README.md (cdd08d3738d03fa1c5b01be4db3c08aa805025cd)


Co-authored-by: Uppari Gopi <GopiUppari@users.noreply.huggingface.co>
---
 README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/README.md b/README.md
index 8ea750a..fb18818 100644
--- a/README.md
+++ b/README.md
@@ -74,6 +74,7 @@ With instruction-tuned models, you need to use chat templates to process our inp
 
 ```python
 from transformers import pipeline
+import torch
 
 pipe = pipeline("text-generation", model="google/gemma-3-1b-it", device="cuda", torch_dtype=torch.bfloat16)