All integrations
Streamlit

Streamlit + Meridian

Route your Streamlit app's LLM calls through Meridian with a single base URL override. Keep your existing st.session_state patterns — just point the client at Meridian's gateway.

Install

pip install openai streamlit

Configure

import streamlit as st
from openai import OpenAI

if "client" not in st.session_state:
    st.session_state.client = OpenAI(
        api_key="sk-meridian-...",
        base_url="https://api.meridian.sh/v1",
    )

Use

response = st.session_state.client.chat.completions.create(
    model="meridian-4",
    messages=[{"role": "user", "content": prompt}],
)
st.write(response.choices[0].message.content)

Session state is scoped to Meridian

The client lives in st.session_state so it persists across reruns. All requests flow through Meridian's gateway — rate limits, logging, and key rotation are handled upstream.