### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### The Feature

Add ChatGPT Subscription (`chatgpt/`) provider model entries for the newly released GPT-5.4 family models that are available through ChatGPT subscriptions:

chatgpt/gpt-5.4-mini — GPT-5.4 mini chatgpt/gpt-5.4-mini-fast — GPT-5.4 mini with Fast Mode chatgpt/gpt-5.4-fast — GPT-5.4 with Fast Mode

### Motivation, pitch

I’m using the `chatgpt/` provider with ChatGPT Subscription authentication, and I’d like LiteLLM to support the newer GPT-5.4 model variants exposed through ChatGPT subscriptions.

Currently, I would like to use these models through the `chatgpt/` provider:

`chatgpt/gpt-5.4-fast` `chatgpt/gpt-5.4-mini` `chatgpt/gpt-5.4-mini-fast`

This is mainly useful for downstream tools such as Codex CLI and OpenCode, where users may want: - a faster GPT-5.4 option - a lower-cost GPT-5.4 mini option - a fast mode variant for GPT-5.4 mini as well

Related issues: - `anomalyco/opencode#16499` — Request for GPT-5.4 fast mode support in OpenCode

### What part of LiteLLM is this about?

Proxy

### LiteLLM is hiring a founding backend engineer, are you…