Node.js services, on your ChatGPT subscription.
Use the OpenAI SDK you already know. Change the base URL, authenticate with a ProxyLLM key, and your service's OpenAI calls run through Codex Hosted instead of metered API billing.
$129/month SaaS. Bring your own model keys. No inference markup.
Three steps to connect.
Create a ProxyLLM key
Generate a scoped key in the dashboard. Give each service its own key and budget cap so agent loops and background jobs stay bounded.
Point the OpenAI SDK at ProxyLLM
Set baseURL to https://api.proxyllm.ai/v1 and use your ProxyLLM key. The official OpenAI Node.js SDK keeps the same chat completions shape.
Let Codex Hosted take the call
OpenAI model names run through Codex Hosted on your connected ChatGPT account. If you hit a plan limit, ProxyLLM falls back to your own OpenAI key until the window resets.
One OpenAI client.
The SDK call stays normal. ProxyLLM sits behind the OpenAI-compatible base URL.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.proxyllm.ai/v1",
apiKey: process.env.PROXYLLM_API_KEY,
});
const result = await client.chat.completions.create({
model: "gpt-4o-mini",
messages: [{ role: "user", content: "Summarize this support thread." }],
});
console.log(result.choices[0]?.message?.content);Run your AI workloads on your ChatGPT subscription.
ProxyLLM runs OpenAI's Codex for you, signed in with your own ChatGPT account. Your apps call one OpenAI-compatible endpoint and the work bills to your flat plan instead of per-token API pricing.
Put a budget cap on every service.
Create separate keys for workers, web apps, and cron jobs, then follow every request in the cost dashboard.