sushruthsam/codechat

1import streamlit as st2from ctransformers import AutoModelForCausalLM3 4# Load the model5llm = AutoModelForCausalLM.from_pretrained(6    model_path_or_repo_id="mistral-7b-instruct-v0.1.Q2_K.gguf",7    model_type="mistral",8)9 10st.title("Conversational Chat with Mistral 🦙🗨️")11 12 13# Function to generate response14def generate_response(user_query):15    prompt = f"""The user query is {user_query} """16    args = {17        "prompt": prompt,18        "stream": True,19        "max_new_tokens": 4096,20        "temperature": 0,21    }22 23    response_placeholder = st.empty()  # Placeholder for displaying response chunks24 25    response_so_far = ""  # Initialize empty string to store cumulative response26 27    for chunk in llm(**args):28        response_so_far += chunk  # Append current chunk to cumulative response29        response_placeholder.write(response_so_far)  # Display cumulative response30 31    return  # No need to return anything32 33 34# User input35user_query = st.text_input("Enter your query:", "")36 37if user_query:38    # Generate and display response39    generate_response(user_query)40