### Check for existing issues

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

### What happened?

When a user with role internal_user or internal_user_viewer has their budget exceeded, calling `GET /v1/models `or `GET /models `returns a 400 budget_exceeded error instead of the model list.

A proxy admin with an exhausted budget is unaffected — they can still list models.

Expected: 200 OK with the model list. Budget enforcement should only block inference routes (/chat/completions, /completions, etc.), not read-only model discovery.

### Steps to Reproduce

1. Create an internal user with 0 budget 2. Generate a key for that user 3. Try hitting these endpoints `GET /v1/models `or `GET /models ` Observed: 400 budget_exceeded

Expected: 200 OK with model list

### Relevant log output

```shell

```

### What part of LiteLLM is this about?

Proxy

### What LiteLLM version are you on ?

v1.83.14-stable

### Twitter / LinkedIn details

_No response_