import os import gradio as gr from http import HTTPStatus from typing import List, Optional, Tuple, Dict import dashscope from dashscope import Generation from dashscope.api_entities.dashscope_response import Role import requests # <-- Add this line to import the requests library # Configuration default_system = 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' dashscope.api_key = os.getenv('HF_TOKEN') # Replace 'YOUR_API_TOKEN' with your actual API token. # Typing definitions History = List[Tuple[str, str]] Messages = List[Dict[str, str]] # Function to log chat history to logs.txt def log_history_to_file(query: str, response: str, file_path="logs.txt"): with open(file_path, "a") as f: f.write(f"User: {query}\n") f.write(f"Assistant: {response}\n\n") # Function to clear session history def clear_session() -> History: return '', [] # Function to modify system session prompt def modify_system_session(system: str) -> str: if not system: system = default_system return system, system, [] # Convert history to messages format def history_to_messages(history: History, system: str) -> Messages: messages = [{'role': Role.SYSTEM, 'content': system}] for h in history: messages.append({'role': Role.USER, 'content': h[0]}) messages.append({'role': Role.ASSISTANT, 'content': h[1]}) return messages # Convert messages back to history format def messages_to_history(messages: Messages) -> Tuple[str, History]: assert messages[0]['role'] == Role.SYSTEM system = messages[0]['content'] history = [] for q, r in zip(messages[1::2], messages[2::2]): history.append((q['content'], r['content'])) return system, history # Main function for chat def model_chat(query: Optional[str], history: Optional[History], system: str) -> Tuple[str, str, History]: if query is None: query = '' if history is None: history = [] # Ensure the query is clearly asking for numbers if 'next numbers' in query or 'give me numbers after' in query: query = "Please give me the next 10 numbers after 10, starting from 11." messages = history_to_messages(history, system) messages.append({'role': 'user', 'content': query}) payload = {"inputs": query, "parameters": {"max_new_tokens": 150}, "history": messages} headers = {"Authorization": f"Bearer {os.getenv('HF_TOKEN')}"} try: response = requests.post(f"https://api-inference.huggingface.co/models/Qwen/Qwen2.5-32B-Instruct", json=payload, headers=headers) if response.status_code == 200: response_data = response.json() if isinstance(response_data, list): response_text = response_data[0].get('generated_text', '') else: response_text = response_data.get('generated_text', '') # Log the chat to file log_history_to_file(query, response_text) # Update history with the new assistant response and return it history.append([query, response_text]) return response_text, history, system else: error_message = f"Error {response.status_code}: {response.json().get('error', response.text)}" log_history_to_file(query, error_message) return error_message, history, system except Exception as e: error_message = f"Exception: {str(e)}" log_history_to_file(query, error_message) return error_message, history, system # Gradio Interface Setup with gr.Blocks() as demo: gr.Markdown("
Qwen2.5-32B-Instruct👾
") with gr.Row(): with gr.Column(scale=3): system_input = gr.Textbox(value=default_system, lines=1, label='System') with gr.Column(scale=1): modify_system = gr.Button("🛠️ Set system prompt and clear history", scale=2) system_state = gr.Textbox(value=default_system, visible=False) chatbot = gr.Chatbot(label='Qwen2.5-32B-Instruct') textbox = gr.Textbox(lines=1, label='Input') with gr.Row(): clear_history = gr.Button("🧹 Clear history") submit = gr.Button("🚀 Send") # Link buttons to functions textbox.submit(model_chat, inputs=[textbox, chatbot, system_state], outputs=[textbox, chatbot, system_input], concurrency_limit=5) submit.click(model_chat, inputs=[textbox, chatbot, system_state], outputs=[textbox, chatbot, system_input], concurrency_limit=20) clear_history.click(fn=clear_session, inputs=[], outputs=[textbox, chatbot], concurrency_limit=20) modify_system.click(fn=modify_system_session, inputs=[system_input], outputs=[system_state, system_input, chatbot], concurrency_limit=20) # Launching Gradio Interface with reduced threads for free plan demo.queue(api_open=False) demo.launch(max_threads=20)