Skip to content
Summer sale. 50% off Pro, $19.50 for life. Desktop app: Apple Silicon Mac only.SUMMER50
CodexUse

Using the Codex App without an OpenAI subscription

GuideUpdated August 2026 · 5 min read

You can run the official Codex App on a model that is not OpenAI’s, using a provider you pay directly. Some of the app stops working when you do, and it is better to know which parts before you set it up.

Short answer: chat, tool calls, and reasoning effort work on a custom-provider profile. ChatGPT-hosted features do not, because there is no ChatGPT account behind that profile. You still need the Codex App installed.

What works

  • Ordinary conversation with the model you configured.
  • Tool calls and their continuations, including multi-step work.
  • Reasoning effort, where your model supports it, with the values it actually advertises.
  • Restarting the window and resuming a thread.
  • Separate history and settings, because the profile has its own Codex home.

What does not work

  • Anything that calls ChatGPT’s own backend: dictation, browser remote control, and similar hosted features. They require a ChatGPT account, and a custom-provider profile does not have one.
  • OpenAI models on that profile. If you want those, use an OpenAI profile.
  • Providers that do not speak the OpenAI Responses or Chat Completions shape.

The honest cost picture

You are trading a subscription for metered provider billing. Whether that is cheaper depends entirely on your volume and the model you pick, and it can go either way. If you use the app heavily on a frontier model, a subscription may well be the better deal. If you use it moderately, or a cheaper model is good enough for most of your work, metered billing usually wins. Measure your own usage before assuming.

What this does not claim

  • CodexUse is not affiliated with OpenAI, and this is not a way around OpenAI billing. A custom-provider profile does not call OpenAI at all — it calls the provider you configured, with your own credential, and you pay that provider directly.
  • A profile is either an OpenAI account or a custom provider. It is never both. Your ChatGPT profiles keep using ChatGPT exactly as before.
  • Hosted ChatGPT features do not work on a custom-provider profile — dictation, browser remote control, and anything that calls ChatGPT’s own backend need a ChatGPT account. Chat, tool calls, and reasoning effort do work.
  • Your provider must speak the OpenAI Responses or Chat Completions shape. Providers with a bespoke protocol are not supported.
  • macOS on Apple Silicon only. Intel Mac, Windows, and Linux desktop builds are not in the current release line.
  • You are responsible for the third-party provider’s terms and costs, and for reviewing your own OpenAI plan terms.

Related

Docs: Custom model providers Guide: run a custom model provider Is running a custom provider allowed? Use an aggregator like OpenRouter Reasoning effort for custom models

Do I still need the Codex App itself?

Yes. CodexUse drives the official app; it does not replace it. You need the app installed on macOS, but a custom-provider profile does not need a ChatGPT subscription to run its own turns.

Can I mix a subscription and a custom provider?

Across profiles, yes — run OpenAI profiles and custom-provider profiles side by side. Within one profile, no. A profile is one or the other by design.

Is this cheaper?

Sometimes. You replace a flat subscription with metered provider billing, so it depends on your volume and your model choice. We would not claim a saving without knowing your usage.