Comfy-Org/ComfyUI · error · NotImplementedError

activation incorrectly specified. check the config file and

Error message

activation incorrectly specified. check the config file and look for 'activation'.

What it means

Raised by the BigVGAN AMPBlock1 (labeled line in mmaudio/vae/bigvgan.py) when h.activation is neither 'snake' nor 'snakebeta'. The activation name comes from the VAE hyperparameter config object h; any typo or unsupported activation aborts VAE construction with NotImplementedError.

Source

Thrown at comfy/ldm/mmaudio/vae/bigvgan.py:81

                       padding=get_padding(kernel_size, 1))
        ])

        self.num_layers = len(self.convs1) + len(self.convs2)  # total number of conv layers

        if activation == 'snake':  # periodic nonlinearity with snake function and anti-aliasing
            self.activations = nn.ModuleList([
                Activation1d(
                    activation=activations.Snake(channels, alpha_logscale=h.snake_logscale))
                for _ in range(self.num_layers)
            ])
        elif activation == 'snakebeta':  # periodic nonlinearity with snakebeta function and anti-aliasing
            self.activations = nn.ModuleList([
                Activation1d(
                    activation=activations.SnakeBeta(channels, alpha_logscale=h.snake_logscale))
                for _ in range(self.num_layers)
            ])
        else:
            raise NotImplementedError(
                "activation incorrectly specified. check the config file and look for 'activation'."
            )

    def forward(self, x):
        acts1, acts2 = self.activations[::2], self.activations[1::2]
        for c1, c2, a1, a2 in zip(self.convs1, self.convs2, acts1, acts2):
            xt = a1(x)
            xt = c1(xt)
            xt = a2(xt)
            xt = c2(xt)
            x = xt + x

        return x


class AMPBlock2(torch.nn.Module):

    def __init__(self, h, channels, kernel_size=3, dilation=(1, 3), activation=None):

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Set h.activation to 'snake' or 'snakebeta' (lowercase, exact)
2. If you edited the config, restore the original from the checkpoint
3. Check for whitespace/case in the config string

Example fix

# before
h.activation = 'Snake'
# after
h.activation = 'snake'

if h.activation not in ('snake', 'snakebeta'):
    raise ValueError(f"activation must be 'snake' or 'snakebeta', got {h.activation!r}")

def is_supported_activation(a: str) -> bool:
    return a in ('snake', 'snakebeta')

When it happens

Trigger: Building the MMAudio VAE with h.activation set to something like 'Snake', 'relu', 'mish', or missing so it falls to the else branch.

Common situations: Hand-editing the VAE config JSON; porting a BigVGAN config that uses a case variant or an activation this fork does not implement.

Related errors

• Unknown activation type: {activation_type}
• Minimum cutoff must be larger than zero.
• A cutoff above 0.5 does not make sense.
• Unknown model: {name}
• Unknown normalization type: {norm_type}

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/8e1ad5f960ecae77. Report an issue: GitHub.