missing "transcribe" in ModelType

[yuming-long]

Describe the bug
got pydantic error

    validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self)
pydantic_core._pydantic_core.ValidationError: 1 validation error for ModelObject
type
  Input should be 'chat', 'language', 'code', 'image', 'embedding', 'moderation', 'rerank' or 'audio' [type=enum, input_value='transcribe', input_type=str]
    For further information visit https://errors.pydantic.dev/2.11/v/enum

To Reproduce

from together import Together

client = Together(api_key="XXX")
client.models.list()

I am using latest version

uv pip show together
Name: together
Version: 1.5.18
Location: venv/lib/python3.13/site-packages
Requires: aiohttp, click, eval-type-backport, filelock, numpy, pillow, pydantic, requests, rich, tabulate, tqdm, typer
Required-by:

Expected behavior
I want be able to call client.models.list() without error

Additional context
I can see there is only one model openai/whisper-large-v3 that only has type transcribe