import gradio as gr from llama_cpp import Llama from huggingface_hub import hf_hub_download # Fetch the model file from Hugging Face Hub model_repo = "SupermanRX/moderateTherapistModel" # Replace with your repo name model_file = "model.gguf" # Replace with your GGUF model file name in the repo # Download the model file to the local environment model_path = hf_hub_download(repo_id=model_repo, filename=model_file) # Load the GGUF model llm = Llama(model_path=model_path) # Define the chatbot function def chatbot(input_text): output = llm(prompt=input_text, max_tokens=200) return output['choices'][0]['text'] # Create Gradio interface with gr.Blocks() as demo: chatbot_ui = gr.Chatbot() textbox = gr.Textbox(label="Type your message here:") submit = gr.Button("Send") # Handle user interaction def user_interaction(input_text, chat_history): response = chatbot(input_text) chat_history.append((input_text, response)) return chat_history, "" submit.click(user_interaction, [textbox, chatbot_ui], [chatbot_ui, textbox]) demo.launch()