Comfy-Org/ComfyUI · error · ValueError

unknown merge strategy {self.merge_strategy}

Error message

unknown merge strategy {self.merge_strategy}

What it means

Raised by the VideoAnimationMerge/alpha-mixing module (util.py) when merge_strategy is neither 'fixed', 'learned', nor 'learned_with_images'. An assert just above already constrains membership in self.strategies, so this else-branch ValueError is the hard backstop for a strategy string outside the known set — it fires at module init.

Source

Thrown at comfy/ldm/modules/diffusionmodules/util.py:47

        super().__init__()
        self.merge_strategy = merge_strategy
        self.rearrange_pattern = rearrange_pattern

        assert (
            merge_strategy in self.strategies
        ), f"merge_strategy needs to be in {self.strategies}"

        if self.merge_strategy == "fixed":
            self.register_buffer("mix_factor", torch.Tensor([alpha]))
        elif (
            self.merge_strategy == "learned"
            or self.merge_strategy == "learned_with_images"
        ):
            self.register_parameter(
                "mix_factor", torch.nn.Parameter(torch.Tensor([alpha]))
            )
        else:
            raise ValueError(f"unknown merge strategy {self.merge_strategy}")

    def get_alpha(self, image_only_indicator: torch.Tensor, device) -> torch.Tensor:
        # skip_time_mix = rearrange(repeat(skip_time_mix, 'b -> (b t) () () ()', t=t), '(b t) 1 ... -> b 1 t ...', t=t)
        if self.merge_strategy == "fixed":
            # make shape compatible
            # alpha = repeat(self.mix_factor, '1 -> b () t  () ()', t=t, b=bs)
            alpha = self.mix_factor.to(device)
        elif self.merge_strategy == "learned":
            alpha = torch.sigmoid(self.mix_factor.to(device))
            # make shape compatible
            # alpha = repeat(alpha, '1 -> s () ()', s = t * bs)
        elif self.merge_strategy == "learned_with_images":
            if image_only_indicator is None:
                alpha = rearrange(torch.sigmoid(self.mix_factor.to(device)), "... -> ... 1")
            else:
                alpha = torch.where(
                    image_only_indicator.bool(),
                    torch.ones(1, 1, device=image_only_indicator.device),

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Set merge_strategy to 'fixed' (uses the constant alpha buffer) or 'learned'/'learned_with_images'
2. Restore the original model config from the checkpoint
3. Check for case sensitivity and stray whitespace in the config string

Example fix

# before
merge_strategy='Fixed'
# after
merge_strategy='fixed'

if merge_strategy not in ('fixed', 'learned', 'learned_with_images'):
    raise ValueError(f"merge_strategy must be fixed|learned|learned_with_images, got {merge_strategy!r}")

def is_valid_merge_strategy(s: str) -> bool:
    return s in ('fixed', 'learned', 'learned_with_images')

When it happens

Trigger: Constructing the time-mixing module with merge_strategy set to a typo ('Learned'), an unsupported value ('interpolated'), or None in a video-model config.

Common situations: Loading animated/video diffusion model configs (AnimateDiff-style) edited by hand; configs from forks that define extra strategies not present in this build.

Related errors

• ar_video sampler requires 5-D video latents [B,C,T,H,W], got
• Unknown rope_img: {rope_img}
• Unsupported spatial_scale {scale}. Choose from {list(mapping
• Invalid {cls.__name__} string: '{value}'. Valid values are:
• Invalid normalization type: {normtype}

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/144e0b086e4b7da8. Report an issue: GitHub.