import gradio as gr from huggingface_hub import InferenceClient import os api_key=os.environ.get('qwen_API_KEY') """ 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("Qwen/Qwen2.5-72B-Instruct",token=api_key) 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]}) messages.append({"role": "user", "content": message}) response = "" 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 response += token yield response # Gradio ChatInterface setup with MathJax mathjax_script = """ """ demo = gr.ChatInterface( respond, examples=[["你好,你是谁?"], ["你是谁开发的?"]], cache_examples=False, title="千问2.5-72B", description="千问2.5-72B聊天机器人", additional_inputs=[ gr.Textbox(value="You are Qwen, created by Alibaba Cloud. You are a helpful assistant.", label="System message"), gr.Slider(minimum=1, maximum=8888, value=2048, 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)"), ], css=mathjax_script, chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="panel") ) if __name__ == "__main__": demo.launch()