invoke-ai/InvokeAI · error · RuntimeError

Error scanning the model at {path} for malware. Aborting imp

Error message

Error scanning the model at {path} for malware. Aborting import.

What it means

read_checkpoint_meta() treats a picklescan error (scan_result.scan_err) — meaning the scan itself failed and safety is unknown — as fatal. Unless picklescan is disabled via config, the import is aborted with this RuntimeError.

Source

Thrown at invokeai/backend/model_manager/util/model_util.py:80

    else:
        if scan:
            scan_result = pscan.scan_file_path(path)
            if scan_result.infected_files != 0:
                if get_config().unsafe_disable_picklescan:
                    logger.warning(
                        f"The model {path} is potentially infected by malware, but picklescan is disabled. "
                        "Proceeding with caution."
                    )
                else:
                    raise RuntimeError(f"The model {path} is potentially infected by malware. Aborting import.")
            if scan_result.scan_err:
                if get_config().unsafe_disable_picklescan:
                    logger.warning(
                        f"Error scanning the model at {path} for malware, but picklescan is disabled. "
                        "Proceeding with caution."
                    )
                else:
                    raise RuntimeError(f"Error scanning the model at {path} for malware. Aborting import.")

        checkpoint = torch.load(path, map_location=torch.device("meta"))
    return checkpoint


def lora_token_vector_length(checkpoint: dict[str | int, torch.Tensor]) -> Optional[int]:
    """
    Given a checkpoint in memory, return the lora token vector length

    :param checkpoint: The checkpoint
    """

    def _get_shape_1(key: str, tensor: torch.Tensor, checkpoint: dict[str | int, torch.Tensor]) -> Optional[int]:
        lora_token_vector_length = None

        if "." not in key:
            return lora_token_vector_length  # wrong key format
        model_key, lora_key = key.split(".", 1)

View on GitHub (pinned to 0b6a024f2f)

Solutions

1. Re-download the model (file may be corrupt) and retry the scan.
2. Run picklescan manually (picklescan --path file.ckpt) to see the underlying scan error.
3. Upgrade/downgrade picklescan to a version compatible with the file.
4. If the source is trusted and the scan is a known false failure, set unsafe_disable_picklescan=true in config.

from picklescan.scanner import scan_file_path
try:
    r = scan_file_path(model_file)
    if r.scan_err:
        print(f'picklescan could not analyze {model_file}: investigate before import')
except Exception as e:
    print(f'scan failed: {e}')

try:
    ckpt = read_checkpoint_meta(path)
except RuntimeError as e:
    if 'Error scanning the model' in str(e):
        run_picklescan_manually(path)  # diagnose; re-download if corrupt
    else:
        raise

When it happens

Trigger: picklescan raises or returns scan_err=True while scanning a checkpoint during import, e.g. on malformed, encrypted, or unusually structured pickles, or when picklescan encounters an opcode it cannot analyze.

Common situations: Corrupted or truncated downloads; checkpoints saved with exotic pickle protocols or non-standard serialization; picklescan version incompatibilities with new torch pickle formats.

Related errors

• The model {path.stem} is potentially infected by malware. Ab
• Error scanning the model at {path.stem} for malware. Abortin
• The model {path} is potentially infected by malware. Abortin
• only relative download paths accepted
• Cannot derive a safe filename for {url} from '{file_name}'

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/318ab745edb98d11. Report an issue: GitHub.