|
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 |
|
|
|
|
|
|
|
mathjax_script = """ |
|
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.13.11/katex.min.css" integrity="sha384-Um5gpz1odJg5Z4HAmzptBBvMrqERW5Z4icqS+8iu7r5vL20nmpEQ6/6BwYIuL8" crossorigin="anonymous"> |
|
<script defer src="https://cdnjs.cloudflare.com/ajax/libs/KaTeX/0.13.11/katex.min.js" integrity="sha384-YNHdsYkH6gFf146yyNCToHNauALlYwViumE/lp77cA3E2g8VoECF1Sn4aa/n9I" crossorigin="anonymous"></script> |
|
<script defer src="https://cdn.jsdelivr.net/npm/katex@0.13.11/dist/contrib/auto-render.min.js" integrity="sha384-vZTG03m+2yp6N6BNi5iM4rW4oIwk5DfcNdFfxkk9ZWpDriOkXX8vo6LrT9HH" crossorigin="anonymous" |
|
onload="renderMathInElement(document.body);"></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 |
|
) |
|
|
|
if __name__ == "__main__": |
|
demo.launch() |
