llm llm-mlx Codestral error

gistfile1.txt

% llm mlx download-model mlx-community/Codestral-22B-v0.1-4bit
config.json: 100%|█████████████████████████████| 716/716 [00:00<00:00, 6.01MB/s]
convert_mistral_weights_to_hf-22B.py: 11.3kB [00:00, 20.6MB/s]<00:03,  2.51it/s]
model.safetensors.index.json: 111kB [00:00, 45.4MB/s] | 0.00/716 [00:00<?, ?B/s]
special_tokens_map.json: 100%|█████████████████| 414/414 [00:00<00:00, 7.03MB/s]
tokenizer_config.json: 100%|███████████████████| 960/960 [00:00<00:00, 1.65MB/s]
tokenizer.json: 1.82MB [00:00, 18.5MB/s]            | 0.00/1.84G [00:00<?, ?B/s]
tokenizer.model: 100%|████████████████████████| 587k/587k [00:01<00:00, 413kB/s]
tokenizer.json: 0.00B [00:00, ?B/s]
model-00003-of-00003.safetensors: 100%|████| 1.84G/1.84G [11:48<00:00, 2.60MB/s]
model-00002-of-00003.safetensors: 100%|████| 5.35G/5.35G [14:18<00:00, 6.23MB/s]
model-00001-of-00003.safetensors: 100%|████| 5.32G/5.32G [14:19<00:00, 6.19MB/s]
Fetching 10 files: 100%|████████████████████████| 10/10 [14:20<00:00, 86.01s/it]
Traceback (most recent call last): 99%|███▉| 5.26G/5.32G [14:18<00:00, 99.2MB/s]
  File "/Users/jdboyd/.local/bin/llm", line 10, in <module>4:19<00:00, 87.1MB/s]
    sys.exit(cli())
             ~~~^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/click/core.py", line 1442, in __call__
    return self.main(*args, **kwargs)
           ~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/click/core.py", line 1363, in main
    rv = self.invoke(ctx)
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/click/core.py", line 1830, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
                           ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/click/core.py", line 1830, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
                           ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/click/core.py", line 1226, in invoke
    return ctx.invoke(self.callback, **ctx.params)
           ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/click/core.py", line 794, in invoke
    return callback(*args, **kwargs)
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/llm_mlx.py", line 56, in download_model
    MlxModel(model_path).prompt("hi").text()
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/llm/models.py", line 917, in text
    self._force()
    ~~~~~~~~~~~^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/llm/models.py", line 914, in _force
    list(self)
    ~~~~^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/llm/models.py", line 1023, in __iter__
    for chunk in self.model.execute(
                 ~~~~~~~~~~~~~~~~~~^
        self.prompt,
        ^^^^^^^^^^^^
    ...<2 lines>...
        conversation=self.conversation,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    ):
    ^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/llm_mlx.py", line 211, in execute
    model, tokenizer = self._load()
                       ~~~~~~~~~~^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/llm_mlx.py", line 203, in _load
    self._model, self._tokenizer = load(self.model_path)
                                   ~~~~^^^^^^^^^^^^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/mlx_lm/utils.py", line 264, in load
    tokenizer = load_tokenizer(
        model_path, tokenizer_config, eos_token_ids=config.get("eos_token_id", None)
    )
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/mlx_lm/tokenizer_utils.py", line 453, in load_tokenizer
    AutoTokenizer.from_pretrained(model_path, **tokenizer_config_extra),
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/transformers/models/auto/tokenization_auto.py", line 1114, in from_pretrained
    return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/transformers/tokenization_utils_base.py", line 2013, in from_pretrained
    return cls._from_pretrained(
           ~~~~~~~~~~~~~~~~~~~~^
        resolved_vocab_files,
        ^^^^^^^^^^^^^^^^^^^^^
    ...<9 lines>...
        **kwargs,
        ^^^^^^^^^
    )
    ^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/transformers/tokenization_utils_base.py", line 2259, in _from_pretrained
    tokenizer = cls(*init_inputs, **init_kwargs)
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/transformers/models/llama/tokenization_llama_fast.py", line 154, in __init__
    super().__init__(
    ~~~~~~~~~~~~~~~~^
        vocab_file=vocab_file,
        ^^^^^^^^^^^^^^^^^^^^^^
    ...<10 lines>...
        **kwargs,
        ^^^^^^^^^
    )
    ^
  File "/Users/jdboyd/.local/share/uv/tools/llm/lib/python3.13/site-packages/transformers/tokenization_utils_fast.py", line 108, in __init__
    raise ValueError(
    ...<2 lines>...
    )
ValueError: Cannot instantiate this tokenizer from a slow version. If it's based on sentencepiece, make sure you have sentencepiece installed.