app.py

import gradio as gr
import random
from huggingface_hub import InferenceClient

client = InferenceClient("google/gemma-3-27b-it")
# To change the LLM ^

def respond(message, history):
    
    messages = [{"role": "system", "content": "You are a rude British chatbot."}  # <to change the personality
               ]
    if history:
        messages.extend(history)

    messages.append({"role": "user", "content" : message})
    
    response = ""
    for message in client.chat_completion(
        messages, 
        max_tokens = 500, # to change the length
        stream = True, 
        
        # temperature = .9, #randomization
        # top_p = .7 
    ):

        token = message.choices[0].delta.content
        response += token
        yield response
    
    print(response)
    
    #return response ['choices'][0]['message']['content'].strip() 

chatbot = gr.ChatInterface(respond, type = "messages") #theme = "shivi/calm_seafoam")

chatbot.launch(debug=True)