Use Codex CLI with any model

Set it up

1. 01

   The configuration surface

   The file is ~/.codex/config.toml, with a project-scoped .codex/config.toml alongside it. Add a provider and select it:

   ~/.codex/config.toml

   model_provider = "hopscotch"
   model = "unirouter/auto"
   
   [model_providers.hopscotch]
   name = "Hopscotch"
   base_url = "https://api.hopscotchlabs.ai/v1"
   env_key = "HOPSCOTCH_API_KEY"
   wire_api = "responses"

   bash

   export HOPSCOTCH_API_KEY=ub_live_XXXXXXXXXXXX

   base_url must include /v1. Codex appends only /responses to whatever you give it, so a base URL that stops at the hostname posts to /responses and receives a 404.

   Do not name the provider openai. The built-in provider ids are reserved and a block of that name is ignored rather than refused, which is the kind of failure that costs an afternoon. Any other id works; hopscotch above is arbitrary.

2. 02

   The key comes from a variable you name

   env_key names the environment variable Codex reads, and it reads it at request time rather than caching it.

   Codex also has a login flow that writes a credential to ~/.codex/auth.json. That path belongs to the built-in OpenAI provider and is not how a custom provider is authenticated; leave it alone and use env_key.

3. 03

   Pin the model id

   Codex's model field is free text and this API needs it to name a provider, so the default has to change before the first request.

   There is one more model setting, review_model, which /review uses and which defaults to the session model. If you set it, pin it the same way.

These models support tool calling and take at least 128K tokens of context. Prices are per million tokens.

What changes when Codex CLI runs through Hopscotch