KeyError: 'yi' on mtbench

I was trying to mtbench Nous-Capybara-34B-GGUF which has the following config.json:

{
    "model_type": "yi"
}

and I get the following error:

Traceback (most recent call last):
  File "/workspace/FastChat/fastchat/llm_judge/gen_model_answer.py", line 288, in <module>
    run_eval(
  File "/workspace/FastChat/fastchat/llm_judge/gen_model_answer.py", line 55, in run_eval
    get_answers_func(
  File "/workspace/venv_fastchat/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "/workspace/FastChat/fastchat/llm_judge/gen_model_answer.py", line 86, in get_model_answers
    model, tokenizer = load_model(
  File "/workspace/FastChat/fastchat/model/model_adapter.py", line 337, in load_model
    model, tokenizer = adapter.load_model(model_path, kwargs)
  File "/workspace/FastChat/fastchat/model/model_adapter.py", line 73, in load_model
    tokenizer = AutoTokenizer.from_pretrained(
  File "/workspace/venv_fastchat/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py", line 752, in from_pretrained
    config = AutoConfig.from_pretrained(
  File "/workspace/venv_fastchat/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1098, in from_pretrained
    config_class = CONFIG_MAPPING[config_dict["model_type"]]
  File "/workspace/venv_fastchat/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 795, in __getitem__
    raise KeyError(key)
KeyError: 'yi'

which points towards:

https://github.com/huggingface/transformers/blob/3cefac1d974db5e2825a0cb2b842883a628be7a0/src/transformers/models/auto/configuration_auto.py#L30

being the issue, so is this a transformers upstream issue or can yi be mtbenched even with no upstream support? thanks

Reproduce the failure with the linked Yi model and trace the loading path through fastchat/llm_judge/gen_model_answer.py:55-86 and fastchat/model/model_adapter.py:73,337, using the transformers traceback as the entry point. Confirm whether the failure is confined to the model_type lookup; done means the issue is either routed upstream with evidence or Yi loads for mtbench without this KeyError.