app.py

import streamlit as st
from sentence_transformers import SentenceTransformer

# Load model
@st.cache_resource
def load_model():
    return SentenceTransformer("BAAI/bge-small-en")

model = load_model()

# UI
st.title("Text Embedder (BAAI/bge-small-en)")
st.markdown("Enter some text and get the embedding vector.")

# Input
user_input = st.text_area("Text to Embed", height=150)

if st.button("Generate Embedding"):
    if user_input.strip():
        with st.spinner("Embedding..."):
            embedding = model.encode([user_input])[0]
        st.success("Embedding generated!")
        st.json(embedding.tolist())
    else:
        st.warning("Please enter some text.")