From 5a8f502db5889873ffa13132b603b7b6daac605a Mon Sep 17 00:00:00 2001
From: comfyanonymous <121283862+comfyanonymous@users.noreply.github.com>
Date: Tue, 19 Aug 2025 22:08:11 -0700
Subject: [PATCH] Disable prompt weights for qwen. (#9438)

---
 comfy/sd1_clip.py                 | 5 ++++-
 comfy/text_encoders/qwen_image.py | 4 ++--
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/comfy/sd1_clip.py b/comfy/sd1_clip.py
index 1e8adbe69..f8a7c2a1b 100644
--- a/comfy/sd1_clip.py
+++ b/comfy/sd1_clip.py
@@ -534,7 +534,10 @@ class SDTokenizer:
         min_padding = tokenizer_options.get("{}_min_padding".format(self.embedding_key), self.min_padding)
 
         text = escape_important(text)
-        parsed_weights = token_weights(text, 1.0)
+        if kwargs.get("disable_weights", False):
+            parsed_weights = [(text, 1.0)]
+        else:
+            parsed_weights = token_weights(text, 1.0)
 
         # tokenize words
         tokens = []
diff --git a/comfy/text_encoders/qwen_image.py b/comfy/text_encoders/qwen_image.py
index f07318d6c..6646b1003 100644
--- a/comfy/text_encoders/qwen_image.py
+++ b/comfy/text_encoders/qwen_image.py
@@ -15,7 +15,7 @@ class QwenImageTokenizer(sd1_clip.SD1Tokenizer):
     def __init__(self, embedding_directory=None, tokenizer_data={}):
         super().__init__(embedding_directory=embedding_directory, tokenizer_data=tokenizer_data, name="qwen25_7b", tokenizer=Qwen25_7BVLITokenizer)
         self.llama_template = "<|im_start|>system\nDescribe the image by detailing the color, shape, size, texture, quantity, text, spatial relationships of the objects and background:<|im_end|>\n<|im_start|>user\n{}<|im_end|>\n<|im_start|>assistant\n"
-        self.llama_template_images = "<|im_start|>system\nDescribe the key features of the input image \\(color, shape, size, texture, objects, background\\), then explain how the user's text instruction should alter or modify the image. Generate a new image that meets the user's requirements while maintaining consistency with the original input where appropriate.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>{}<|im_end|>\n<|im_start|>assistant\n"
+        self.llama_template_images = "<|im_start|>system\nDescribe the key features of the input image (color, shape, size, texture, objects, background), then explain how the user's text instruction should alter or modify the image. Generate a new image that meets the user's requirements while maintaining consistency with the original input where appropriate.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>{}<|im_end|>\n<|im_start|>assistant\n"
 
     def tokenize_with_weights(self, text, return_word_ids=False, llama_template=None, images=[], **kwargs):
         if llama_template is None:
@@ -25,7 +25,7 @@ class QwenImageTokenizer(sd1_clip.SD1Tokenizer):
                 llama_text = self.llama_template.format(text)
         else:
             llama_text = llama_template.format(text)
-        tokens = super().tokenize_with_weights(llama_text, return_word_ids=return_word_ids, **kwargs)
+        tokens = super().tokenize_with_weights(llama_text, return_word_ids=return_word_ids, disable_weights=True, **kwargs)
         key_name = next(iter(tokens))
         embed_count = 0
         qwen_tokens = tokens[key_name]