CodeX Deployment Guide

OpenAI's enterprise-grade AI coding assistant with intelligent coding and code generation right in your terminal.

💡 Prerequisites

• Node.js 18+ runtime (Windows Install | macOS Install | Linux Install)
• TokenSwitch account with a valid CodeX token (Create Key)
• Recommended: CC-Switch configuration tool

CC-Switch Quick Config (Recommended)

Step 1: Install CodeX

npm install -g @openai/codex@latest

Verify the installation:

codex --version

Step 2: CC-Switch Configuration

1. Open the CC-Switch configuration tool
2. Switch to the Codex tab
3. Click Add Provider and fill in the following:

Setting	Value
Name	TokenSwitch (or a custom name)
Base URL	https://tokenswitch.cc/v1
API Key	Your CodeX token
Model	gpt-5.4

4. Save the configuration and you're ready to go

Manual Installation and Configuration

Windows

Install CodeX

npm install -g @openai/codex@latest

Create Configuration Files

1. Create the configuration directory:

mkdir "$env:USERPROFILE\.codex"

2. Create the config.toml file (path: %USERPROFILE%\.codex\config.toml):

model_provider = "openai"
model = "gpt-5.4"
base_url = "https://tokenswitch.cc/v1"
wire_api = "responses"

3. Create the auth.json file (path: %USERPROFILE%\.codex\auth.json):

{
  "OPENAI_API_KEY": "your-codex-token"
}

macOS

Install CodeX

npm install -g @openai/codex@latest

Create Configuration Files

1. Create the configuration directory:

mkdir -p ~/.codex

2. Create ~/.codex/config.toml:

model_provider = "openai"
model = "gpt-5.4"
base_url = "https://tokenswitch.cc/v1"
wire_api = "responses"

3. Create ~/.codex/auth.json:

{
  "OPENAI_API_KEY": "your-codex-token"
}

Linux

Install CodeX

npm install -g @openai/codex@latest

Create Configuration Files

1. Create the configuration directory:

mkdir -p ~/.codex

2. Create ~/.codex/config.toml:

model_provider = "openai"
model = "gpt-5.4"
base_url = "https://tokenswitch.cc/v1"
wire_api = "responses"

3. Create ~/.codex/auth.json:

{
  "OPENAI_API_KEY": "your-codex-token"
}

Configuration Reference

config.toml Parameters

Parameter	Description	Value
model_provider	Model provider	openai
model	Model to use	gpt-5.4
base_url	API endpoint URL	https://tokenswitch.cc/v1
wire_api	API protocol format	responses

auth.json Parameters

Parameter	Description
OPENAI_API_KEY	Your dedicated CodeX token

Verify Installation

Once configured, run the following in your terminal:

codex

If the CodeX interactive interface launches successfully, deployment is complete.

Troubleshooting

codex: command not found

Possible Cause	Solution
Not installed correctly	Re-run npm install -g @openai/codex@latest
PATH not configured	Check that the npm global bin directory is in your PATH
Terminal not refreshed	Restart the terminal window

Connection Failed / Authentication Error

Possible Cause	Solution
Incorrect API Key	Check that the token in auth.json is correct
Wrong Base URL	Confirm base_url in config.toml is https://tokenswitch.cc/v1
Token group mismatch	Confirm you are using a token group that supports OpenAI models
Insufficient balance	Log in to the TokenSwitch console and check your balance

Configuration Files Not Taking Effect

• Check that file paths are correct: ~/.codex/config.toml and ~/.codex/auth.json
• Check that TOML and JSON formatting is correct (watch for quotes and commas)
• Restart the terminal and try again

Next Steps

✅ CodeX deployment complete!

• CC-Switch Configuration Tool — Graphically manage multiple providers
• FAQ — More troubleshooting help