blakeblackshear/frigate · error · ValueError

Model type "{self.detector_config.model.model_type}" is curr

Error message

Model type "{self.detector_config.model.model_type}" is currently not supported.

What it means

detect_raw dispatched on model_type and the model's model_type is not yologeneric, which is the only post-processing path implemented for the Axengine detector. Any other ModelTypeEnum value in the model metadata (ssd, yolonas, yolox, etc.) reaches the ValueError.

Source

Thrown at frigate/detectors/plugins/axengine.py:95

        return model_props

    def download_model(self, filename):
        if not os.path.isdir(model_cache_dir):
            os.mkdir(model_cache_dir)

        HF_ENDPOINT = os.environ.get("HF_ENDPOINT", "https://huggingface.co")
        urllib.request.urlretrieve(
            f"{HF_ENDPOINT}/AXERA-TECH/frigate-resource/resolve/axmodel/{filename}",
            model_cache_dir + filename,
        )

    def detect_raw(self, tensor_input):
        results = None
        results = self.session.run(None, {"images": tensor_input})
        if self.detector_config.model.model_type == ModelTypeEnum.yologeneric:
            return post_process_yolo(results, self.width, self.height)
        else:
            raise ValueError(
                f'Model type "{self.detector_config.model.model_type}" is currently not supported.'
            )

View on GitHub (pinned to ca18b8dc13)

Solutions

Use a YOLO generic-style model (model_type: yologeneric) with the axengine detector
If a custom model of another architecture is required, verify its model_info type can be set to yologeneric or wait for/patch support in post_process
Double-check the model_info.json / cached metadata for the model actually loaded

Defensive patterns

Strategy: validation

Validate before calling

from frigate.detectors.detector_config import ModelTypeEnum
assert config.model.model_type == ModelTypeEnum.yologeneric, 'axengine only supports yologeneric'

Try / catch

try:
    dets = detector.detect_raw(tensor)
except ValueError as e:
    if 'not supported' in str(e):
        logger.error('Model type incompatible with axengine post-processing')
    raise

Prevention

Use yologeneric models with the axengine detector
Keep model metadata consistent with the actual architecture
Test with a single frame at startup to fail fast

When it happens

Trigger: Running the axengine detector with a custom model whose model_info/model_type is set (or defaults) to anything other than yologeneric, then an inference arrives and detect_raw runs.

Common situations: User supplies a custom .axmodel converted from an SSD or YOLO-NAS model; model_info.json incorrectly labels the type; plugin/version drift where older model metadata uses a different enum value.

Related errors

AI-assisted analysis of blakeblackshear/frigate@ca18b8dc13 (2026-08-27). Data as JSON: /api/errors/fad2ffa52e659ff9. Report an issue: GitHub.