Comfy-Org/ComfyUI · error · ValueError

This Uni3C ControlNet expects a Wan model with dim {}, the l

Error message

This Uni3C ControlNet expects a Wan model with dim {}, the loaded model has dim {}.

What it means

The Uni3C ControlNet's block width (controlnet_blocks[0].norm1.linear.in_features) must equal the target model's dim. A dimensional mismatch means the controlnet was trained for a different Wan variant (e.g. 14B vs 1.3B), and patching would fail or corrupt attention math, so the node reports both dims.

Source

Thrown at comfy_extras/nodes_model_patch.py:733

                              "strength": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
                              "start_percent": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                              "end_percent": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                              }}
    RETURN_TYPES = ("MODEL",)
    FUNCTION = "apply_patch"
    EXPERIMENTAL = True

    CATEGORY = "model/patch/wan"

    def apply_patch(self, model, model_patch, vae, render_video, strength, start_percent, end_percent):
        if not isinstance(model_patch.model, comfy.ldm.wan.uni3c.WanUni3CControlnet):
            raise ValueError("The connected model patch is not a Uni3C ControlNet.")
        cnet_dim = model_patch.model.controlnet_blocks[0].norm1.linear.in_features
        model_dim = getattr(model.get_model_object("diffusion_model"), "dim", None)
        if model_dim is None:
            raise ValueError("The Uni3C ControlNet only works with Wan models.")
        if model_dim != cnet_dim:
            raise ValueError("This Uni3C ControlNet expects a Wan model with dim {}, the loaded model has dim {}.".format(cnet_dim, model_dim))

        model_patched = model.clone()
        model_sampling = model.get_model_object("model_sampling")
        sigma_start = model_sampling.percent_to_sigma(start_percent)
        sigma_end = model_sampling.percent_to_sigma(end_percent)
        latent_format = model.get_model_object("latent_format")
        patch = WanUni3CCnetPatch(model_patch, render_video[:, :, :, :3], vae, latent_format, strength, sigma_start, sigma_end)
        model_patched.set_model_double_block_patch(patch)
        return (model_patched,)


class UsoStyleProjectorPatch:
    def __init__(self, model_patch, encoded_image):
        self.model_patch = model_patch
        self.encoded_image = encoded_image

    def __call__(self, kwargs):
        txt_ids = kwargs.get("txt_ids")

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Download the Uni3C ControlNet variant that matches your Wan model's dim (the error message states both expected and actual).
2. Switch the base model to the variant the controlnet was trained on.
3. Confirm dims in Python: model.get_model_object('diffusion_model').dim vs controlnet_blocks[0].norm1.linear.in_features.

cnet_dim = model_patch.model.controlnet_blocks[0].norm1.linear.in_features
model_dim = model.get_model_object('diffusion_model').dim
if model_dim != cnet_dim:
    raise UserFacingError(f'dim mismatch: controlnet {cnet_dim} vs model {model_dim}')

When it happens

Trigger: Pairing a Uni3C controlnet trained on Wan 1.3B (dim 1536) with a 14B model (dim 5120) or vice versa; any combination where cnet_dim != model_dim.

Common situations: Downloading the controlnet matching the wrong base model size; mixing T2V and I2V variants of different scales; upgrading the base model without re-fetching the controlnet.

Related errors

• The connected model patch is not a Uni3C ControlNet.
• The Uni3C ControlNet only works with Wan models.
• Control type {max_type_name}({max_type}) is out of range for
• y is None, did you try using a controlnet for SDXL on SD1?
• This Controlnet needs a VAE but none was provided, please us

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/f409cf5aafe311a5. Report an issue: GitHub.