app.py

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

# Load the model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("sambanovasystems/SambaLingo-Hungarian-Chat", use_fast=False)
model = AutoModelForCausalLM.from_pretrained("sambanovasystems/SambaLingo-Hungarian-Chat", device_map="auto", torch_dtype="auto")

# Create the pipeline
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, device_map="auto", use_fast=False)

# Define the chat function
def chat(question):
    messages = [{"role": "user", "content": question}]
    prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, truncation=True)
    outputs = pipe(prompt, max_length=1024, truncation=True)[0]
    return outputs["generated_text"]

# Set up the Gradio interface
iface = gr.Interface(
    fn=chat,
    inputs=gr.Textbox(lines=2, placeholder="Kérdezz valamit..."),
    outputs="text",
    title="Teszt Chatbot",
    description="SambaLingo-Hungarian-Chat model"
)

# Launch the interface
iface.launch()