OpenAI models, unchanged. The bill, very changed.
Keep your SDK, prompts, and model names exactly as they are. Swap the base URL and your OpenAI traffic runs on the flat ChatGPT subscription you already pay for, through Codex Hosted. Your API key stays connected as fallback.
$129/month SaaS. Bring your own model keys. No inference markup.
Three steps to connect.
Connect your ChatGPT account
Sign in with OpenAI's official device-code login. ProxyLLM runs Codex in an isolated container for your account, so OpenAI workloads bill to your ChatGPT plan instead of per-token pricing.
Point your SDK
Set the base URL to https://api.proxyllm.ai/v1 and authenticate with your ProxyLLM key. Python, Node, curl: anything that speaks the OpenAI API works unchanged.
Request models as usual
gpt-5, gpt-4o, gpt-4o-mini: same names you use today. Requests run through Codex Hosted. At a plan limit they fall back to a second Codex account or your own OpenAI key, stored AES-256-GCM encrypted.
One client change.
Swap the base URL and key. Everything else in your codebase stays put.
from openai import OpenAI
client = OpenAI(
base_url="https://api.proxyllm.ai/v1",
api_key="pk_live_…", # your ProxyLLM key
)
r = client.chat.completions.create(
model="gpt-4o-mini",
messages=[{"role": "user", "content": "Classify this ticket."}],
)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.
Connect OpenAI in two minutes.
$129/month flat, no inference markup. OpenAI traffic moves off the per-token meter and onto your existing ChatGPT plan.