Codex is OpenAI's command-line coding assistant. It can talk to any custom model provider through its config.toml, so pointing that provider at CrossModel gives you access to the whole catalog.

Quickstart

1. Install Codex

macOS / Linux:

curl -fsSL https://chatgpt.com/codex/install.sh | sh

Windows:

powershell -ExecutionPolicy ByPass -c "irm https://chatgpt.com/codex/install.ps1 | iex"

Or install with npm:

npm install -g @openai/codex

2. Create an API key

Sign in to the Console and create a key that starts with cm- on the API Keys page. The key is shown only once, so copy it somewhere safe.

3. Add CrossModel as a provider

Edit ~/.codex/config.toml and define the CrossModel provider:

model_provider = "crossmodel"
model = "openai/gpt-5.5"
 
[model_providers.crossmodel]
name = "CrossModel"
base_url = "https://api.crossmodel.ai/v1"
env_key = "CROSSMODEL_API_KEY"

4. Set the API key environment variable

env_key tells Codex which environment variable holds your key. Add it to your shell profile (~/.zshrc or ~/.bashrc):

export CROSSMODEL_API_KEY="cm-YOUR_KEY"

Reopen the terminal, or run source ~/.zshrc, to load it.

5. Launch

Run Codex from your project directory:

cd /path/to/your/project
codex

How it works

model_provider selects the default provider, and the [model_providers.crossmodel] block describes it: base_url points at CrossModel's OpenAI-compatible endpoint (https://api.crossmodel.ai/v1), and env_key tells Codex where to read the key. Requests go to CrossModel with your cm- key sent as a Bearer token.

Choosing models

To switch models, just change the model field in config.toml — no reinstall needed. Model IDs follow the vendor/short_name format, e.g. openai/gpt-5-mini or deepseek/deepseek-v4-pro. See the full list in the model catalog.

Troubleshooting