Skip to main content
To facilitate integrating models with the Codex CLI, it is recommended to use the codex-MiniMax-M2.1 model. While models can interact with files using tools like Bash, this approach is not considered optimal for Agent workflows. For better Agent-based practices, we suggest using platforms such as Claude Code or Cursor.

​
Install Codex CLI

Use npm to install the Codex CLI globally: 
npm i -g @openai/codex

​
Configure MiniMax API

Important: Clear OpenAI Environment Variables Before ConfigurationBefore configuring, ensure you clear the following OpenAI-related environment variables to avoid conflicts with MiniMax API:
  • OPENAI_API_KEY
  • OPENAI_BASE_URL
  1. Add the following configuration to the .codex/config.toml file.
  • The base_url should be set based on your location: for international users, use https://api.minimax.io/v1; for users in China, use https://api.minimaxi.com/v1
[model_providers.minimax]
name = "MiniMax Chat Completions API"
base_url = "https://api.minimax.io/v1"
env_key = "MINIMAX_API_KEY"
wire_api = "chat"
requires_openai_auth = false
request_max_retries = 4
stream_max_retries = 10
stream_idle_timeout_ms = 300000

[profiles.m21]
model = "codex-MiniMax-M2.1"
model_provider = "minimax"
  1. Set the API key using environment variables in the current terminal session for security reasons. Use the API key obtained from the MiniMax Developer Platform (For users in China, visit MiniMax Developer Platform) as the value for MINIMAX_API_KEY.
export MINIMAX_API_KEY="<MINIMAX_API_KEY>"
  1. Start the Codex CLI with the specified profile.
codex --profile m21