app .py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline

# Use a public, non-gated model to avoid auth issues
model_id = "tiiuae/falcon-rw-1b"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

generator = pipeline("text-generation", model=model, tokenizer=tokenizer)

def chat_with_cael(prompt):
    result = generator(prompt, max_length=200, do_sample=True, top_k=50)[0]["generated_text"]
    return result

iface = gr.Interface(fn=chat_with_cael, 
                     inputs=gr.Textbox(lines=2, placeholder="Talk to Cael..."),
                     outputs="text",
                     title="Cael: Your AI Companion",
                     description="This is a simple early version of Cael using Falcon-RW-1B. Fully free and open!")

iface.launch()