app.py

import streamlit as st 
import requests
import os

api_token = os.environ.get("api_token")
API_URL = "https://api-inference.huggingface.co/models/google/gemma-7b"
headers = {"Authorization": f"Bearer {api_token}"}
st.title("Google Gemma 7B Chat")
st.write("New powerful text generation model by Google AI trained on 7B parameters. Enter your message and get response in few seconds!")
mainc = st.container(height=600)
cont = mainc.container(height=400)
prompt = mainc.chat_input(placeholder="Eg. Why astronaut riding a horse is most popular prompt for testing ai models?")

def query(payload):
	response = requests.post(API_URL, headers=headers, json=payload)
	return response.json()

if prompt:
    if len(prompt) > 8000:
        errormsg = st.chat_message("Assistant", avatar="⚠")
        errormsg.markdown(":red[Sorry, prompt can't be longer than 8000 tokens!]")
    else:
        output = query({
        	"inputs": prompt,
        })
        if output != None:
            user_msg = cont.chat_message("User")
            user_msg.write(prompt)
            as_msg = cont.chat_message("Assistant")
            as_msg.write(output[0]['generated_text'])