Fix

app.py

@@ -198,12 +198,13 @@ with Blocks() as app:
                 nickname = Textbox(value="用户1234", label="nickname", info="用户昵称")
             json_messages = JSON([], max_height=125, label="JSON消息格式")
 
-    chatbot.clear(
-        lambda: ("[]", ""),
-        outputs=[json_messages, text_response],
-        api_name=False,
-        show_api=False,
-    )
+    ## NOT SUPPORT IN GRADIO 5.0.1
+    # chatbot.clear(
+    #     lambda: ("[]", ""),
+    #     outputs=[json_messages, text_response],
+    #     api_name=False,
+    #     show_api=False,
+    # )
     chatbot.retry(
         callback,
         [textbox, chatbot, json_messages],

requirements.txt

@@ -3,5 +3,5 @@ torchvision
 transformers
 accelerate
 qwen-vl-utils
-gradio==5.13.0
+gradio
 https://github.com/Dao-AILab/flash-attention/releases/download/v2.7.3/flash_attn-2.7.3+cu12torch2.5cxx11abiFALSE-cp310-cp310-linux_x86_64.whl