How to Use Codex on Your Phone for Free – A Step‑by‑Step Guide
This guide shows how OpenAI has embedded Codex into the ChatGPT mobile app, turning your phone into a remote control panel for a desktop‑running Codex instance, and walks through the free setup, Ollama installation, model switching, and connection workflow.
Integration Overview
OpenAI merged Codex into the ChatGPT mobile app. The phone does not run the Codex engine; it only displays a control panel while the engine executes on a host machine (laptop, Mac mini, devbox, or remote environment). Both the Free and Go subscription plans provide access at no extra cost.
Only macOS desktop clients can be paired; Windows clients are not supported. The Codex CLI and IDE plugins are not included in the mobile pairing workflow.
Supported Interaction Features
View live state : The phone shows the live state of all task threads running on the connected machine.
Approve and release : When Codex pauses for confirmation, tapping the phone resumes execution.
See results : Screenshots, terminal output, diffs, and test results are pushed to the phone in real time.
Change direction : New commands or strategy adjustments can be issued from the phone.
Switch model : The underlying GPT model version can be changed directly from the phone.
Create new task : A new thread can be started from the phone without returning to the computer.
Setup Procedure
Upgrade mobile app : Update ChatGPT; the Codex icon appears in the top‑left corner.
Start Codex on desktop : Open the macOS Codex app, click Start in Settings, and grant permission.
Authorize on phone : Authorize the connection; subsequent connections can be re‑established from the same screen.
Complete connection : Desktop Settings → Connection shows the phone’s status; the phone displays the identical Codex view.
Install Ollama : curl -fsSL https://ollama.com/install.sh | sh Launch Codex via Ollama : ollama launch codex-app Select a model : Cloud models such as K2.6 require a subscription. The author used the free Gemma‑4 model, which has a weekly quota.
Switch model : Edit the Codex configuration file to replace the model identifier; the change takes effect without reconnecting the phone.
Undo Ollama configuration : Comment out the added lines in the config file to stop using Ollama.
Extend free usage : After a Plus subscription expires, waiting one week before renewal allows continued use of GPT‑5.5 for free.
Configuration Details
Changing the model is performed by modifying the config file referenced by Codex; the modification is reflected immediately on the phone without a new connection.
Commenting out the Ollama‑related entries in the same config file disables the Ollama backend.
Visual References
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Old Zhang's AI Learning
AI practitioner specializing in large-model evaluation and on-premise deployment, agents, AI programming, Vibe Coding, general AI, and broader tech trends, with daily original technical articles.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
