Skip to main content

Using Codex

Codex can be configured to use the RCD LLM Service through a custom provider in ~/.codex/config.toml.

info

Codex should use the base URL https://llm.rcd.clemson.edu/v1. The chat-completions compatibility endpoint is useful for some other clients, but Codex should be configured with the base URL.

warning

Codex depends on the service's POST /v1/responses compatibility layer. The RCD LLM Service currently provides this through a shim, so Codex support is functional but may change more often than the chat-completions endpoint.

1. Create An API Key​

First, request an allocation, sign in to the service, and create an API key:

2. Export The API Key​

Store the key in an environment variable before starting Codex:

export RCD_LLM_API_KEY="your-api-key-here"

If you want this to be available in new shells automatically, add the same line to your shell startup file such as ~/.zshrc.

3. Choose A Model​

Check the Available Models page for the exact model name to use.

4. Update ~/.codex/config.toml​

Add a provider like the following:

model = "<model-name>"
model_provider = "rcd-llm"
web_search = "disabled"

[model_providers.rcd-llm]
name = "RCD LLM Service"
base_url = "https://llm.rcd.clemson.edu/v1"
env_key = "RCD_LLM_API_KEY"

Replace <model-name> with an exact model name from the Models page.

info

The RCD LLM Service does not support server-side tools like web_search. They must be disabled for things to function correctly.

tip

If you already have model, model_provider, or model_providers entries in your config.toml, merge this snippet into your existing file instead of creating duplicate keys.

5. Start A New Codex Session​

After saving ~/.codex/config.toml, start a new Codex session:

codex

Switching Between OpenAI and RCD LLM Service​

Codex CLI now supports profiles, allowing you to switch between the OpenAI backend and the RCD LLM Service more easily. To do this, add both an openai profile and an rcd-llm profile, then set profile to whichever one you want to use by default. For example, to make rcd-llm the default, use:

profile = "rcd-llm"

[profiles.openai]
model = "gpt-5.4"

[profiles.rcd-llm]
model_provider = "rcd-llm"
web_search = "disabled"
model = "glm-5.1-fp8"

[model_providers.rcd-llm]
name = "RCD LLM Service"
base_url = "https://llm.rcd.clemson.edu/v1"
env_key = "RCD_LLM_API_KEY"

Then, if you run codex, it will use the RCD LLM Service by default, but you can run codex -p openai to use the OpenAI backend.

Model Catalog​

We're piloting an endpoint that automatically generates a Codex-compatible model catalog. This allows you to use the /model command in Codex to switch models and provides some extra information to Codex about model capabilities (context lengths, input modalities, etc.).

To use this, download the catalog and save it as ~/.codex/rcd_llms_model_catalog.json. Then update the model_catalog_json option in ~/.codex/config.toml. For example, if using profiles, use this as the profile:

[profiles.rcd-llm]
model_provider = "rcd-llm"
web_search = "disabled"
model = "glm-5.1-fp8"
model_catalog_json = "~/.codex/rcd_llms_model_catalog.json"
warning

The model catalog is a snapshot of available models. It will not update if we change our available models. You will have to download the catalog again to update.

Notes​

  • You must still be on the Clemson network or connected through VPN.
  • If the service model list changes, update the model value in ~/.codex/config.toml.
  • Keep the API key out of Git repositories, shared shell scripts, and support tickets.