CodeX Deployment Guide
Enterprise-grade AI coding assistant deployment guide
Prerequisites
Please complete the Node.js environment installation and CC-Switch tool installation first.
CodeX is a next-generation intelligent programming assistant based on the GPT-5 architecture, providing developers with exceptional code generation and optimization capabilities.
Deployment Path: System environment setup ➜ CLI tool installation ➜ API integration ➜ Development environment ready
Quick Configuration with CC-Switch (Recommended)
Prerequisite
Before configuring CodeX with CC-Switch, ensure you have globally installed the CodeX tool via npm:
npm install -g @openai/codex@latestVerify installation: codex --version
We recommend using the CC-Switch configuration tool for graphical configuration — simple and quick with no command line needed.
Configuration Steps
Launch CC-Switch and Switch to the Codex Tab
- Open the CC-Switch application
- Click the "Codex" tab at the top
- Click the orange "+" button in the upper right corner to add a new configuration
Fill in CodeX Provider Configuration
- Provider Name: Custom name (e.g., "元渊智能")
- Base URL: Enter
https://api.ai.soraliststudio.com/v1 - API Key: Paste your CodeX-specific token from the 元渊智能 platform (codex test token group)
- Model: Select
gpt-5-codex - Other settings: Adjust reasoning intensity, network access, etc., as needed
- Click "Save"
- CC-Switch will automatically create
~/.codex/config.tomlandauth.jsonfiles - You can add multiple provider configurations and switch between them anytime
- After switching, close and restart CodeX for changes to take effect
Enable Configuration
- Find the "元渊智能" configuration you just created
- Click the "Currently Used" button on the right side
- The configuration will be marked as "Currently Used" (green label)
- Restart CodeX for the new configuration to take effect
System Tray Quick Switch
CC-Switch supports quick CodeX configuration switching via the system tray:
- Right-click the CC-Switch icon in the system tray
- Select the Codex category from the menu
- Directly select the configuration to switch to
- Configuration takes effect immediately
- Make sure to create a dedicated key from the "codex test" token group on the 元渊智能 platform
- CodeX tokens are not interchangeable with Claude Code tokens
- You need to restart CodeX after switching configurations
- You can test API endpoint speed in CC-Switch
Manual Command Line Configuration
If you don't use CC-Switch, follow these steps to manually configure CodeX.
Step 1: Deploy the CodeX CLI Tool
Launch Command Prompt or PowerShell with administrator privileges:
npm install -g @openai/codex@latest
codex --versionStep 2: Integrate 元渊智能 API Service
Obtain your API credentials:
- Visit the 元渊智能 developer console
- Register or log in to your account
- Navigate to the "API Key Management" module
- When creating a new key, select the "codex test" token group
- Securely save the generated API Key
Security Reminder
CodeX requires an independent token group configuration, completely isolated from the Claude Code token system.
Build the configuration directory:
mkdir %USERPROFILE%\.codex
cd %USERPROFILE%\.codexCreate the configuration file config.toml:
model_provider = "yyzn"
model = "gpt-5.2-codex"
model_reasoning_effort = "xhigh"
network_access = "enabled"
disable_response_storage = true
approval_policy = "on-request"
web_search = "live"
sandbox_mode = "danger-full-access"
collaboration_mode = true
model_supports_reasoning_summaries = true
[model_providers.yyzn]
name = "yyzn"
base_url = "https://api.ai.soraliststudio.com/v1"
wire_api = "responses"
requires_openai_auth = true
[features]
shell_snapshot = true
collab = false
collaboration_modes = false
steer = false
unified_exec = falseCreate the authentication file auth.json:
{
"OPENAI_API_KEY": "Paste your CodeX-specific token here"
}Step 3: Initialize Workspace
mkdir my-codex-project
cd my-codex-project
codexDeploy CodeX Tool
npm install -g @openai/codex@latest
codex --versionIntegrate API Service
Create configuration directory:
mkdir -p ~/.codex
cd ~/.codexCreate config.toml:
cat > config.toml << 'EOF'
model_provider = "yyzn"
model = "gpt-5.2-codex"
model_reasoning_effort = "xhigh"
network_access = "enabled"
disable_response_storage = true
approval_policy = "on-request"
web_search = "live"
sandbox_mode = "danger-full-access"
collaboration_mode = true
model_supports_reasoning_summaries = true
[model_providers.yyzn]
name = "yyzn"
base_url = "https://api.ai.soraliststudio.com/v1"
wire_api = "responses"
requires_openai_auth = true
[features]
shell_snapshot = true
collab = false
collaboration_modes = false
steer = false
unified_exec = false
EOFCreate auth.json:
cat > auth.json << 'EOF'
{
"OPENAI_API_KEY": "Paste your CodeX-specific token here"
}
EOFInitialize Workspace
mkdir my-codex-project
cd my-codex-project
codexDeploy CodeX Tool
sudo npm install -g @openai/codex@latest
codex --versionIntegrate API Service
Create configuration directory:
mkdir -p ~/.codex
cd ~/.codexCreate config.toml:
cat > config.toml << 'EOF'
model_provider = "yyzn"
model = "gpt-5.2-codex"
model_reasoning_effort = "xhigh"
network_access = "enabled"
disable_response_storage = true
approval_policy = "on-request"
web_search = "live"
sandbox_mode = "danger-full-access"
collaboration_mode = true
model_supports_reasoning_summaries = true
[model_providers.yyzn]
name = "yyzn"
base_url = "https://api.ai.soraliststudio.com/v1"
wire_api = "responses"
requires_openai_auth = true
[features]
shell_snapshot = true
collab = false
collaboration_modes = false
steer = false
unified_exec = false
EOFCreate auth.json:
cat > auth.json << 'EOF'
{
"OPENAI_API_KEY": "Paste your CodeX-specific token here"
}
EOFInitialize Workspace
mkdir my-codex-project
cd my-codex-project
codexFAQ
Are CodeX and Claude Code tokens interchangeable?
No, they use different token groups:
- Claude Code: Uses the Claude Code token group
- CodeX: Uses the "codex test" token group
Please create dedicated tokens on the 元渊智能 platform.
Where are configuration files located?
- Windows:
%USERPROFILE%\.codex\ - macOS/Linux:
~/.codex/
How is this guide?