How to load models from a local Windows path?

[3dluvr]

I am trying to load existing models i have from a different folder (portable ComfyUI installation) instead of your models/ local to DSS - how do I that?

I provided the paths as:

model_manager = ModelManager(device="cpu")
model_manager.load_models(
    [
        "G:\\ComfyUI_windows_portable\\ComfyUI\\models\\diffusion_models\\Wan2_1-T2V-1_3B_bf16.safetensors",
        "G:\\ComfyUI_windows_portable\\ComfyUI\\models\\clip\\umt5-xxl-enc-fp8_e4m3fn.safetensors",
        "G:\\ComfyUI_windows_portable\\ComfyUI\\models\\vae\\Wan2_1_VAE_fp32.safetensors",
    ],
    torch_dtype=torch.float8_e4m3fn,
)

...and by all accounts it should just work. While it appears at first that it would load them, it fails and reports an error:

Loading models from: G:\ComfyUI_windows_portable\ComfyUI\models\diffusion_models\Wan2_1-T2V-1_3B_bf16.safetensors
    model_name: wan_video_dit model_class: WanModel
        This model is initialized with extra kwargs: {'model_type': 't2v', 'patch_size': (1, 2, 2), 'text_len': 512, 'in_dim': 16, 'dim': 1536, 'ffn_dim': 8960, 'freq_dim': 256, 'text_dim': 4096, 'out_dim': 16, 'num_heads': 12, 'num_layers': 30, 'window_size': (-1, -1), 'qk_norm': True, 'cross_attn_norm': True, 'eps': 1e-06}
    The following models are loaded: ['wan_video_dit'].
Loading models from: G:\ComfyUI_windows_portable\ComfyUI\models\clip\umt5-xxl-enc-fp8_e4m3fn.safetensors
    model_name: wan_video_text_encoder model_class: WanTextEncoder
    The following models are loaded: ['wan_video_text_encoder'].
Loading models from: G:\ComfyUI_windows_portable\ComfyUI\models\vae\Wan2_1_VAE_fp32.safetensors
    model_name: wan_video_vae model_class: WanVideoVAE
    The following models are loaded: ['wan_video_vae'].
Using wan_video_text_encoder from G:\ComfyUI_windows_portable\ComfyUI\models\clip\umt5-xxl-enc-fp8_e4m3fn.safetensors.
Traceback (most recent call last):
  File "G:\FluxGym\Lib\site-packages\transformers\utils\hub.py", line 403, in cached_file
    resolved_file = hf_hub_download(
                    ^^^^^^^^^^^^^^^^
  File "G:\FluxGym\Lib\site-packages\huggingface_hub\utils\_validators.py", line 106, in _inner_fn
    validate_repo_id(arg_value)
  File "G:\FluxGym\Lib\site-packages\huggingface_hub\utils\_validators.py", line 160, in validate_repo_id
    raise HFValidationError(
huggingface_hub.errors.HFValidationError: Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are forbidden, '-' and '.' cannot start or end the name, max length is 96: 'G:\ComfyUI_windows_portable\ComfyUI\models\clip\google/umt5-xxl'.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "G:\FluxGym\DiffSynth-Studio\examples\wanvideo\my_wan_1.3b_text_to_video.py", line 19, in <module>
    pipe = WanVideoPipeline.from_model_manager(model_manager, torch_dtype=torch.bfloat16, device="cuda")
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "G:\FluxGym\Lib\site-packages\diffsynth\pipelines\wan_video.py", line 142, in from_model_manager
    pipe.fetch_models(model_manager)
  File "G:\FluxGym\Lib\site-packages\diffsynth\pipelines\wan_video.py", line 131, in fetch_models
    self.prompter.fetch_tokenizer(os.path.join(os.path.dirname(tokenizer_path), "google/umt5-xxl"))
  File "G:\FluxGym\Lib\site-packages\diffsynth\prompters\wan_prompter.py", line 94, in fetch_tokenizer
    self.tokenizer = HuggingfaceTokenizer(name=tokenizer_path, seq_len=self.text_len, clean='whitespace')
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "G:\FluxGym\Lib\site-packages\diffsynth\prompters\wan_prompter.py", line 45, in __init__
    self.tokenizer = AutoTokenizer.from_pretrained(name, **kwargs)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "G:\FluxGym\Lib\site-packages\transformers\models\auto\tokenization_auto.py", line 871, in from_pretrained
    tokenizer_config = get_tokenizer_config(pretrained_model_name_or_path, **kwargs)
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "G:\FluxGym\Lib\site-packages\transformers\models\auto\tokenization_auto.py", line 703, in get_tokenizer_config
    resolved_config_file = cached_file(
                           ^^^^^^^^^^^^
  File "G:\FluxGym\Lib\site-packages\transformers\utils\hub.py", line 469, in cached_file
    raise EnvironmentError(
OSError: Incorrect path_or_model_id: 'G:\ComfyUI_windows_portable\ComfyUI\models\clip\google/umt5-xxl'. Please provide either the path to a local folder or the repo_id of a model on the Hub.

Why is it trying to load the models from HF Hub, why is it even going to the Hub when I provided local model paths??

[Artiprocher]

@3dluvr You need to put the tokenizer files into the same folder.

[3dluvr]

Uh, I swear I should pay more attention to the last line in the traceback and not wander above!

Thanks!