LangChain

LangChain’s OpenAI integration accepts a custom base URL, so fremai plugs into any LangChain app (chains, agents, RAG) with a two-field change.

Python

pip install langchain-openai

import os
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="glm-5.2",
    base_url="https://api.fremai.eu/v1",
    api_key=os.environ["FREMAI_API_KEY"],  # "sk-fremai-…"
)

resp = llm.invoke("Explain EU data sovereignty in two sentences.")
print(resp.content)

Use it anywhere a chat model is expected — for example in a chain:

from langchain_core.prompts import ChatPromptTemplate

prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a concise assistant."),
    ("human", "{question}"),
])
chain = prompt | llm
print(chain.invoke({"question": "What is an embedding?"}).content)

Embeddings

from langchain_openai import OpenAIEmbeddings

embeddings = OpenAIEmbeddings(
    model="<embedding-model-id>",  # see GET /models
    base_url="https://api.fremai.eu/v1",
    api_key=os.environ["FREMAI_API_KEY"],
)
vectors = embeddings.embed_documents(["EU-sovereign inference"])

JavaScript / TypeScript

npm install @langchain/openai

import { ChatOpenAI } from "@langchain/openai";

const llm = new ChatOpenAI({
  model: "glm-5.2",
  apiKey: process.env.FREMAI_API_KEY, // "sk-fremai-…"
  configuration: { baseURL: "https://api.fremai.eu/v1" },
});

const resp = await llm.invoke("Explain EU data sovereignty in two sentences.");
console.log(resp.content);

Notes

Model ids come from the catalog or GET /models.
fremai speaks the OpenAI wire format, so LangChain features that rely on it — streaming, tool calling where the model supports it, structured output — work through the same integration.

curl