app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HIT-TMG/bge-m3_RAG-conversational-IR")
model = AutoModelForCausalLM.from_pretrained("HIT-TMG/bge-m3_RAG-conversational-IR")

def predict(input_text):
    inputs = tokenizer(input_text, return_tensors="pt")
    outputs = model.generate(**inputs, max_length=100)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

iface = gr.Interface(fn=predict, inputs="text", outputs="text")
iface.launch()