Comfy-Org/ComfyUI · error · RuntimeError

ERROR: audio encoder file is invalid and does not contain a

Error message

ERROR: audio encoder file is invalid and does not contain a valid model.

What it means

LoadAudioEncoder loads a torch file from the audio_encoders folder and hands its state dict to load_audio_encoder_from_sd; when no known key signature matches, the factory returns None and this RuntimeError reports that the file is not a recognized audio encoder model.

Source

Thrown at comfy_extras/nodes_audio_encoder.py:30

            node_id="AudioEncoderLoader",
            display_name="Load Audio Encoder",
            category="model/loaders",
            inputs=[
                io.Combo.Input(
                    "audio_encoder_name",
                    options=folder_paths.get_filename_list("audio_encoders"),
                ),
            ],
            outputs=[io.AudioEncoder.Output()],
        )

    @classmethod
    def execute(cls, audio_encoder_name) -> io.NodeOutput:
        audio_encoder_name = folder_paths.get_full_path_or_raise("audio_encoders", audio_encoder_name)
        sd = comfy.utils.load_torch_file(audio_encoder_name, safe_load=True)
        audio_encoder = comfy.audio_encoders.audio_encoders.load_audio_encoder_from_sd(sd)
        if audio_encoder is None:
            raise RuntimeError("ERROR: audio encoder file is invalid and does not contain a valid model.")
        return io.NodeOutput(audio_encoder)


class AudioEncoderEncode(io.ComfyNode):
    @classmethod
    def define_schema(cls) -> io.Schema:
        return io.Schema(
            node_id="AudioEncoderEncode",
            category="model/conditioning",
            inputs=[
                io.AudioEncoder.Input("audio_encoder"),
                io.Audio.Input("audio"),
            ],
            outputs=[io.AudioEncoderOutput.Output()],
        )

    @classmethod
    def execute(cls, audio_encoder, audio) -> io.NodeOutput:

View on GitHub (pinned to 1c6d8d45b3)

Solutions

1. Verify the file is the intended audio encoder model from a supported source and re-download if corrupt
2. Inspect the state dict keys (torch.load / safetensors) and compare against the key signatures load_audio_encoder_from_sd expects
3. Ensure ComfyUI and the encoder file versions match — update one or the other
4. Remove misplaced files from models/audio_encoders so the combo lists only valid encoders

Example fix

// before
# models/audio_encoders/ contains 'some_vae.safetensors' -> selected -> raises

// after
# keep only genuine encoder files, e.g.:
# models/audio_encoders/encodec_v1_5hz.safetensors
sd = comfy.utils.load_torch_file(path, safe_load=True)
print(list(sd)[:10])  # confirm keys look like an audio encoder

sd = comfy.utils.load_torch_file(path, safe_load=True)
from comfy.audio_encoders.audio_encoders import load_audio_encoder_from_sd
if load_audio_encoder_from_sd(sd) is None:
    print('not a valid audio encoder:', list(sd)[:8])  # diagnose keys

try:
    enc = LoadAudioEncoder.execute(name)
except RuntimeError as e:
    if 'invalid and does not contain a valid model' in str(e):
        remove_or_replace_file(name)  # re-download correct encoder
    else:
        raise

When it happens

Trigger: Placing an arbitrary .pt/.safetensors (e.g. a VAE, checkpoint, or corrupt file) in models/audio_encoders and selecting it; a partially downloaded encoder; an encoder saved in a format/key-layout the loader does not recognize (version skew).

Common situations: Wrong file dropped into the audio_encoders directory; updating ComfyUI while keeping an old-format encoder file; filename collisions where the wrong model is picked.

Related errors

• Unrecognized optical flow model format: expected a torchvisi
• INVALID_TAG_FILTER
• ERROR: audio encoder file is invalid or unsupported embed_di
• ERROR: audio encoder not supported.
• Control type {max_type_name}({max_type}) is out of range for

AI-assisted analysis of Comfy-Org/ComfyUI@1c6d8d45b3 (2026-08-14). Data as JSON: /api/errors/3a8750a03ac78403. Report an issue: GitHub.