app.py

import gradio as gr
from huggingface_hub import InferenceClient
with open("knowledge.txt" , "r", encoding="utf-8") as f:
    knowledge_base.txt = f.read()

client = InterferenceClient("google/gemma-3-27b-it")

def respond(message,history):
    messages = [{"role": "system" , "content" : "You're a supportive and helpful feminist"}]
    if history:
        messages.extend(history)
        
    messages.append({"role" : "user", "content" : message})
    
    response = ""
    for message in client.chat_completion(
        messages, 
        max_tokens = 150,
        stream=True,
    ):
        token = message.choices[0].delta.content
        response += token
        yield response 
        
    print(response)
        
    
chatbot = gr.ChatInterface(respond, type = "messages")

chatbot.launch(debug=True)