invoke-ai/InvokeAI · warning · NotAMatchError

external API models are not probed from disk

Error message

external API models are not probed from disk

What it means

External API model configs (models served by remote providers, identified by provider_id + provider_model_id) are created from API listings, never probed from files on disk. Their from_model_on_disk therefore unconditionally raises NotAMatchError. This is expected behavior: a file probe should never resolve to an ExternalApiModelConfig.

Source

Thrown at invokeai/backend/model_manager/configs/external_api.py:113

    source: str = Field(default="")
    hash: str = Field(default="")
    file_size: int = Field(default=0, ge=0)

    model_config = ConfigDict(extra="forbid")

    @model_validator(mode="after")
    def _populate_external_fields(self) -> "ExternalApiModelConfig":
        if not self.path:
            self.path = f"external://{self.provider_id}/{self.provider_model_id}"
        if not self.source:
            self.source = self.path
        if not self.hash:
            self.hash = f"external:{self.provider_id}:{self.provider_model_id}"
        return self

    @classmethod
    def from_model_on_disk(cls, mod: ModelOnDisk, override_fields: dict[str, object]) -> Self:
        raise NotAMatchError("external API models are not probed from disk")

View on GitHub (pinned to 0b6a024f2f)

Solutions

1. Do not include external API configs in disk-probe candidate lists; create them via the provider API sync instead
2. Instantiate external API models directly with provider_id and provider_model_id (hash auto-generated as external:<provider>:<model>)
3. Catch NotAMatchError in probe loops and continue to the next config class — this raise is the intended 'not applicable' signal
4. File a bug if InvokeAI's own installer routes local files to the external-API config

Example fix

// before
cfg = ExternalApiModelConfig.from_model_on_disk(mod, override_fields)
// after
cfg = ExternalApiModelConfig(provider_id="openai", provider_model_id="gpt-image-1")

from invokeai.backend.model_manager.configs.external_api import ExternalApiModelConfig

def should_probe_on_disk(cls) -> bool:
    return not issubclass(cls, ExternalApiModelConfig)

try:
    config = cls.from_model_on_disk(mod, override_fields)
except NotAMatchError:
    continue  # expected for external API configs in a probe loop

When it happens

Trigger: Model-manager probe/install flow iterating config classes and calling from_model_on_disk on a file, reaching the external-API config class; custom code explicitly calling ExternalApiModelConfig.from_model_on_disk with a ModelOnDisk.

Common situations: Writing a custom model-install routine that probes all config subclasses against local files; a bug in registration order causing external API configs to be consulted during disk probing.

Understand the failure class

Background: UnsupportedOperationException and "is not supported" errors: when a library deliberately refuses a call — this error's family across 30 libraries.

Related errors

• External API models cannot be loaded from disk
• Model config discriminator value must be computed from a dic
• The model with key {key} is not a main SD 1/2/XL checkpoint
• Model '{body.model_key}' not found
• Unsupported control_lllite type: {type(control_lllite)}

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/45d5100985f72240. Report an issue: GitHub.