import spaces import gradio as gr import torch import subprocess import requests from gradio import State # Function to start the ochat server @spaces.GPU def start_ochat_server(): print(f"Is CUDA available: {torch.cuda.is_available()}") print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}") command = [ "python", "-m", "ochat.serving.openai_api_server", "--model", "openchat/openchat_3.5" ] # Start the server in a separate process try: subprocess.Popen(command) return "ochat server started successfully" except Exception as e: return f"Failed to start ochat server: {e}" start_ochat_server() def user(message, history): return "", history + [[message, None]] def bot(history): return chat_with_ochat(history[-1][0]), history + [[None, chat_with_ochat(history[-1][0])]] # Function to send a message to the ochat server and get a response def chat_with_ochat(message): url = "http://0.0.0.0:18888/v1/chat/completions" headers = {"Content-Type": "application/json"} data = { "model": "openchat_3.5", "messages": [{"role": "user", "content": message}] } try: response = requests.post(url, json=data, headers=headers) if response.status_code == 200: return response.json()['choices'][0]['message']['content'] else: return f"Error: Server responded with status code {response.status_code}" except requests.RequestException as e: return f"Error: {e}" # Create a Gradio Blocks interface with session state with gr.Blocks(theme=gr.themes.Soft()) as app: gr.Markdown("## vLLM OpenChat-3.5 Interface") gr.Markdown("Run on your own machine using this command: ```docker run -it -p 7860:7860 --platform=linux/amd64 --gpus all \ registry.hf.space/macadeliccc-openchat-3-5-chatbot:latest python app.py```") message = gr.Textbox(label="Your Message", placeholder="Type your message here") chatbot = gr.Chatbot() clear = gr.Button("Clear") chat_history = State([]) # Session state for chat history message.submit(user, [message, chatbot], [message, chatbot], queue=False).then( bot, chatbot, chatbot ) clear.click(lambda: None, None, chatbot, queue=False) app.launch()