Does not work with `mflux`
#1
by
yankexe
- opened
Looks like the mflux team is planning to add support for this model: https://github.com/filipstrand/mflux/issues/308
Following the model card instructions, go this error:
mflux-generate-qwen \
--model mlx-community/Qwen-Image-2512-4bit \
--prompt "A photorealistic cat wearing a tiny top hat" \
--steps 20
Fetching 17 files: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 17/17 [00:00<00:00, 186657.51it/s]
Fetching 2 files: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 4556.55it/s]
Traceback (most recent call last):
File "/Users/xyz/play/mlx-play/.venv/bin/mflux-generate-qwen", line 10, in <module>
sys.exit(main())
^^^^^^
File "/Users/xyz/play/mlx-play/.venv/lib/python3.12/site-packages/mflux/generate_qwen.py", line 27, in main
qwen = QwenImage(
^^^^^^^^^^
File "/Users/xyz/play/mlx-play/.venv/lib/python3.12/site-packages/mflux/models/qwen/variants/txt2img/qwen_image.py", line 39, in __init__
QwenImageInitializer.init(
File "/Users/xyz/play/mlx-play/.venv/lib/python3.12/site-packages/mflux/models/qwen/qwen_initializer.py", line 36, in init
tokenizer_handler = QwenTokenizerHandler(
^^^^^^^^^^^^^^^^^^^^^
File "/Users/xyz/play/mlx-play/.venv/lib/python3.12/site-packages/mflux/models/qwen/tokenizer/qwen_tokenizer_handler.py", line 20, in __init__
self.tokenizer_raw = transformers.Qwen2Tokenizer.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/xyz/play/mlx-play/.venv/lib/python3.12/site-packages/transformers/tokenization_utils_base.py", line 2113, in from_pretrained
return cls._from_pretrained(
^^^^^^^^^^^^^^^^^^^^^
File "/Users/xyz/play/mlx-play/.venv/lib/python3.12/site-packages/transformers/tokenization_utils_base.py", line 2359, in _from_pretrained
tokenizer = cls(*init_inputs, **init_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/xyz/play/mlx-play/.venv/lib/python3.12/site-packages/transformers/models/qwen2/tokenization_qwen2.py", line 172, in __init__
with open(vocab_file, encoding="utf-8") as vocab_handle:
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: expected str, bytes or os.PathLike object, not NoneType