Pity it doesn't support other llms.

It does, it's just a bit annoying.

I have this set up as a shell script (or you could make it an alias):

    codex --config model="gpt-oss-120b" --config model_provider=custom

    [model_providers.custom]
    name = "Llama-swap Local Service"
    base_url = "http://localhost:8080/v1"
    http_headers = { "Authorization" = "Bearer sk-123456789" }
    wire_api = "chat"

    # Default model configuration
    model = "gpt-oss-120b"
    model_provider = "custom"