From 5dfab7d363eddc6bcd0a00546754fbd5ac041113 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 17 Apr 2025 22:03:11 -0700
Subject: [PATCH] New attempt at solving the exl2 issue

---
 modules/text_generation.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/modules/text_generation.py b/modules/text_generation.py
index cbe5b61b..d62441df 100644
--- a/modules/text_generation.py
+++ b/modules/text_generation.py
@@ -263,6 +263,9 @@ def apply_stopping_strings(reply, all_stop_strings):
 
 
 def get_reply_from_output_ids(output_ids, state=None, starting_from=0):
+    if torch.cuda.is_available():
+        torch.cuda.synchronize()
+
     reply = decode(output_ids[starting_from:], state['skip_special_tokens'] if state else True)
 
     # Handle tokenizers that do not add the leading space for the first token