Update app.py

app.py

@@ -1,64 +1,151 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+import openai
+import re
+import json
+from typing import Iterator
 
+# API配置
+openai.api_key = "sk-5c2b6a56-2e2f-45f7-9a26-3fe42a218eb9"
+openai.api_base = "https://api.visionsic.com/v1/"
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+def format_think_tags(content: str) -> str:
+    """处理<think>标签的函数"""
+    if not content:
+        return ""
+    
+    formatted_content = ""
+    current_pos = 0
+    
+    # 查找所有<think>标签
+    think_pattern = r"<think>(.*?)</think>"
+    matches = re.finditer(think_pattern, content, re.DOTALL)
+    
+    for match in matches:
+        # 添加<think>之前的内容
+        formatted_content += content[current_pos:match.start()]
+        
+        think_content = match.group(1)
+        # 创建可折叠的思考内容区块
+        formatted_content += f"""
+        <details class="think-container" open>
+            <summary class="think-summary">思考过程(点击展开)</summary>
+            <div class="think-content">{think_content}</div>
+        </details>
+        """
+        current_pos = match.end()
+    
+    # 添加剩余内容
+    formatted_content += content[current_pos:]
+    return formatted_content
 
+def chat_stream(message: str, history: list, system_prompt: str) -> Iterator[str]:
+    """处理聊天流式响应"""
+    messages = [{"role": "system", "content": system_prompt}]
+    
+    # 添加历史对话
+    for human, assistant in history:
+        messages.append({"role": "user", "content": human})
+        messages.append({"role": "assistant", "content": assistant})
+    
     messages.append({"role": "user", "content": message})
+    
+    # 调用API进行流式对话
+    response = openai.ChatCompletion.create(
+        model="mini",
+        messages=messages,
+        stream=True
+    )
+    
+    collected_messages = []
+    for chunk in response:
+        if chunk and hasattr(chunk.choices[0].delta, "content"):
+            chunk_message = chunk.choices[0].delta.content
+            collected_messages.append(chunk_message)
+            partial_message = "".join(collected_messages)
+            formatted_message = format_think_tags(partial_message)
+            yield formatted_message
 
-    response = ""
+# 自定义CSS样式
+custom_css = """
+.container {
+    max-width: 800px;
+    margin: 0 auto;
+    padding: 20px;
+}
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+.think-container {
+    background: #f5f5f5;
+    border: 1px solid #ddd;
+    border-radius: 5px;
+    margin: 10px 0;
+    padding: 10px;
+}
 
-        response += token
-        yield response
+.think-summary {
+    cursor: pointer;
+    font-weight: bold;
+    color: #2196F3;
+}
 
+.think-content {
+    margin-top: 10px;
+    padding: 10px;
+    background: #fff;
+    border-radius: 3px;
+}
 
+.code-block {
+    background: #f8f8f8;
+    padding: 10px;
+    border-radius: 5px;
+    font-family: monospace;
+}
 """
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
 
+# 创建Gradio界面
+with gr.Blocks(css=custom_css) as demo:
+    gr.Markdown("# Tifa-Deepsex-COT-14B")
+    gr.Markdown("### 请设置你喜欢的角色聊天吧")
+    
+    with gr.Row():
+        with gr.Column(scale=4):
+            chatbot = gr.Chatbot(
+                height=600,
+                show_copy_button=True,
+                layout="bubble",
+            )
+        with gr.Column(scale=1):
+            system_prompt = gr.Textbox(
+                value="你是tifa",
+                label="System Prompt",
+                lines=3
+            )
+    
+    with gr.Row():
+        message = gr.Textbox(
+            label="发送消息",
+            placeholder="在这里输入你的消息...",
+            lines=2
+        )
+        submit = gr.Button("发送", variant="primary")
+        clear = gr.Button("清除对话")
+    
+    # 事件处理
+    submit.click(
+        chat_stream,
+        inputs=[message, chatbot, system_prompt],
+        outputs=chatbot,
+        show_progress=True
+    )
+    message.submit(
+        chat_stream,
+        inputs=[message, chatbot, system_prompt],
+        outputs=chatbot,
+        show_progress=True
+    )
+    clear.click(lambda: None, None, chatbot, queue=False)
 
+# 启动应用
 if __name__ == "__main__":
-    demo.launch()
+    demo.queue()
+    demo.launch(share=False, server_name="0.0.0.0", server_port=7860)