Update app.py

app.py

@@ -15,13 +15,11 @@ if not os.path.exists(model_path):
 # ウェブUIの起動
 import gradio as gr
 import copy
-import time
 from llama_cpp import Llama
 
 llm = Llama(
     model_path=model_path,
     n_ctx=2048,
-    # n_gpu_layers=100, # CPUで実行する場合は削除
 )
 
 history = []
@@ -57,19 +55,20 @@ def generate_text(message, history):
         yield temp
 
     history.append((message, temp))
-
+    return history
 
 demo = gr.ChatInterface(
-    generate_text,
+    fn=lambda message, history: generate_text(message, history),
     title="ZhongJingGPT-V2-1_8B-GGUF chatbot using llama-cpp-python",
-    description="",
-    examples=["我发热，咳嗽，咽痛，该怎么办"],
+    description="This is a TCM medical assistant chatbot.",
+    examples=[("我发热，咳嗽，咽痛，该怎么办",)],
     cache_examples=True,
     retry_btn=None,
     undo_btn="Remove last",
     clear_btn="Clear all",
 )
-demo.queue(concurrency_count=1, max_size=5)
-demo.launch(debug=True, share=True)
+
+demo.queue(concurrency_limit=1)  # 修改为 concurrency_limit
+demo.launch(debug=True, share=True, max_threads=1)  # 添加 max_threads 参数