# Use your ChatGPT subscription via Codex (no API key). Dimmy Help. > Use your ChatGPT subscription via Codex (no API key). Step-by-step help for Dimmy, the native voice dictation app for macOS, Windows, and Linux. Source: https://dimmy.app/help/integrations-codex-cli/ [← Back to Help center](https://dimmy.app/help/) Integrations # Use your ChatGPT subscription via Codex (no API key) If you already pay for ChatGPT Plus, Pro, Team, Business, or Enterprise, route Dimmy's LLM rewrites and recaps through the local codex CLI. No API key, no per-token cost. Dimmy can run OpenAI's official codex CLI as a subprocess and use it as the LLM provider for style rewrites and meeting recaps. The CLI owns your login session (your ChatGPT Plus / Pro / Team / Business / Enterprise plan), and Dimmy only talks to its stdin/stdout. Dimmy never reads your credentials file. info This is the OpenAI sibling of the [Claude subscription via Claude CLI](https://dimmy.app/help/integrations-claude-cli/) feature. Same idea, different vendor: spend your existing ChatGPT plan instead of an OpenAI API key. ## Prerequisites * An active paid ChatGPT plan: Plus, Pro, Team, Business, or Enterprise. A free ChatGPT account won't work for Codex. * The codex CLI installed. It ships as a native binary, so Node.js is optional. * A browser, for the sign-in flow. ## Install the CLI If you have Node.js, the fastest path is npm: bash `npm install -g @openai/codex` No Node.js? Grab the native installer from OpenAI (developers.openai.com/codex/cli). Open a NEW terminal so PATH refreshes, then verify it's there: bash `codex --version` tip After installing on Windows, you may need to restart Dimmy so the new PATH is picked up. The native Windows installer drops codex.exe under %LOCALAPPDATA%\\Programs\\OpenAI\\Codex\\bin and only updates the registry PATH for future shells. Dimmy looks there directly, but a full restart is the sure fix if Recheck still shows "not found". ## Sign in Easiest path: let Dimmy do it. Open the ChatGPT (Codex) card in Dimmy (see below) and click Sign in with ChatGPT. Dimmy launches codex login in a terminal, you complete the ChatGPT login in your browser, and Dimmy detects when it's done. You can also just run codex login in a terminal yourself. info Dimmy spawns the codex CLI and uses its stored login. It never reads \~/.codex/auth.json: that file belongs to the CLI. Dimmy only checks that you're signed in. Use of your ChatGPT plan this way is subject to OpenAI's terms of service. ## Turn it on in Dimmy The connection card lives on Settings → Integrations in the "OpenAI subscription" group. Where you flip the provider on differs slightly per platform. | macOS | Windows | | | --------------- | --------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------- | | Connection card | Settings → Integrations, "OpenAI subscription" group, the "ChatGPT (Codex) subscription" card | Settings → Integrations, "OPENAI SUBSCRIPTION" group, the "ChatGPT (Codex)" card | | Turn it on | On the card, flip "Use ChatGPT (Codex) for recap and rewrite" | Settings → Output → LLM, turn on "Use ChatGPT subscription" | | For recaps | Same toggle covers recaps | Settings → Output → Recap, turn on "Use ChatGPT subscription for recap" | 1. Open the ChatGPT (Codex) card on Settings → Integrations. If nothing is installed yet, click Set up wizard for the guided install + sign-in walkthrough. 2. If it's not signed in, click Sign in with ChatGPT and finish the browser login. 3. Click Test (Test connection on Windows) to send a tiny ping through codex and confirm it answers end-to-end. 4. Flip on the subscription toggle as in the table above. 5. Press your hotkey and dictate a sentence with an LLM style active. The result now comes from your ChatGPT subscription. Settings → Integrations: the ChatGPT (Codex) subscription card showing a green signed-in state Step 1 of 2 · Connect ChatGPT (Codex) wizard warning Linux has no Settings UI for this. The codex routing exists in Dimmy's shared core, but there is no card or wizard on the Linux build, so there's no in-app way to turn it on. This integration is Windows and macOS only today. (Same caveat as [Command Mode](https://dimmy.app/help/command-mode/) and [meeting recaps](https://dimmy.app/help/use-meeting-recaps/).) ## Which model runs When you use the card's toggle, Dimmy points the LLM at codex://default, so Codex uses its own configured default model. If instead you keep the OpenAI provider selected and turn on the subscription toggle, Dimmy passes the OpenAI model you picked to codex. There's no separate Dimmy model dropdown on the Codex card itself. ## What this affects * Style rewrites (Correct, Summarize, Elaborate, Prompt, and friends). Every dictation that passes through an LLM style now uses your subscription. * Meeting recaps, when the subscription toggle is on (one toggle on macOS, the separate recap toggle on Windows). * Custom prompts you defined in [Custom prompts](https://dimmy.app/help/custom-prompts/). They keep working, only the provider changes. ## What this does NOT affect * STT (speech-to-text). That still runs locally (Whisper) or via your chosen STT provider. This integration is LLM-only. * Telemetry. Dimmy records aggregate events (status probed, login spawned, invocation success / failure) with no prompt or response content. See [Telemetry](https://dimmy.app/help/telemetry/). * Other AI tools on your machine. Dimmy spawns its own subprocess; it does not interfere with codex when you use it manually. ## Cost and quotas Dimmy makes one CLI call per LLM action (one rewrite, one recap). Each call counts against your ChatGPT plan's usage. There is no separate Dimmy quota: the CLI talks to OpenAI with your account. info Prefer Anthropic? The same pattern works with your Claude plan. See [Use your Claude subscription via Claude CLI](https://dimmy.app/help/integrations-claude-cli/). Was this helpful? ## In this category [ IntegrationsConnect Claude Desktop to Dimmy (MCP bridge)One-click wizard installs a small MCP server. Ask Claude Desktop about your meetings, dictations, and let it write recaps back into Dimmy.](https://dimmy.app/help/integrations-claude-desktop/) [ IntegrationsUse your Claude subscription via Claude CLI (no API key)If you already pay for Claude Pro, Team, or Max, route Dimmy's LLM rewrites and recaps through the local claude CLI. No API key, no per-token cost.](https://dimmy.app/help/integrations-claude-cli/) [ IntegrationsUse Dimmy's MCP server from Claude CLI (advanced)Dimmy's MCP bridge has a one-click setup for Claude Desktop. There's no built-in button for the terminal, but you can point the Claude CLI at the same binary by hand. Here's how.](https://dimmy.app/help/integrations-mcp-claude-cli/) [ IntegrationsSend meeting recaps to NotionDimmy has a built-in Notion integration. Connect once with your own integration token, pick a destination page or database, then send any recap with one button (or auto-send every meeting).](https://dimmy.app/help/integrations-notion/) [ IntegrationsSend audio from your phone with TelegramConnect Dimmy to your own Telegram account. Record audio on your phone with any app, share it to your Telegram Saved Messages, and Dimmy transcribes it and writes a recap on your computer. Windows and macOS.](https://dimmy.app/help/integrations-telegram/) Last updated: 2026-05-22 On this page 1. [01 Prerequisites](#prerequisites) 2. [02 Install the CLI](#install-the-cli) 3. [03 Sign in](#sign-in) 4. [04 Turn it on in Dimmy](#turn-it-on-in-dimmy) 5. [05 Which model runs](#which-model-runs) 6. [06 What this affects](#what-this-affects) 7. [07 What this does NOT affect](#what-this-does-not-affect) 8. [08 Cost and quotas](#cost-and-quotas)