blakeblackshear/frigate · error · Exception

{self.memx_model_type} is currently not supported for memryx

Error message

{self.memx_model_type} is currently not supported for memryx. See the docs for more info on supported models.

What it means

MemryX detector output post-processing only implements branches for yolonas/ssd/yolox/yologeneric; any other ModelTypeEnum value in outputs dispatch falls through to this Exception at inference time. The configured model type is not supported by the memryx plugin.

Source

Thrown at frigate/detectors/plugins/memryx.py:860

    def process_output(self, *outputs):
        """Output callback function -- receives frames from the MX3 and triggers post-processing"""
        if self.memx_model_type == ModelTypeEnum.yologeneric:
            # Use complete YOLOv9-style postprocessing (includes NMS)
            final_detections = self.post_process_yolo_optimized(outputs)

            self.output_queue.put(final_detections)

        elif self.memx_model_type == ModelTypeEnum.yolonas:
            return self.post_process_yolonas(outputs)

        elif self.memx_model_type == ModelTypeEnum.yolox:
            return self.post_process_yolox(outputs)

        elif self.memx_model_type == ModelTypeEnum.ssd:
            return self.post_process_ssdlite(outputs)

        else:
            raise Exception(
                f"{self.memx_model_type} is currently not supported for memryx. See the docs for more info on supported models."
            )

    def set_stop_event(self, stop_event):
        """Set the stop event for graceful shutdown."""
        self.stop_event = stop_event

    def shutdown(self):
        """Gracefully shutdown the MemryX accelerator"""
        try:
            if hasattr(self, "accl") and self.accl is not None:
                self.accl.shutdown()
                logger.info("MemryX accelerator shutdown complete")
        except Exception as e:
            logger.error(f"Error during MemryX shutdown: {e}")

    def detect_raw(self, tensor_input: np.ndarray):
        """Removed synchronous detect_raw() function so that we only use async"""

View on GitHub (pinned to ca18b8dc13)

Solutions

Use one of the supported model types for memryx (yolonas, ssd, yolox, yologeneric)
Upgrade Frigate to a version whose memryx plugin supports your model type
Re-package/label the model with a supported type if the architecture actually matches one

Defensive patterns

Strategy: validation

Validate before calling

from frigate.detectors.detector_config import ModelTypeEnum
SUPPORTED_MEMRYX = {ModelTypeEnum.yolonas, ModelTypeEnum.ssd, ModelTypeEnum.yolox, ModelTypeEnum.yologeneric}
assert config.model.model_type in SUPPORTED_MEMRYX

Try / catch

try:
    dets = detector.detect_raw(tensor)
except Exception as e:
    if 'not supported for memryx' in str(e):
        logger.error('Use a supported memryx model type')
    raise

Prevention

Restrict memryx models to the documented supported types
Keep model metadata in sync with the actual architecture
Smoke-test a single inference after detector init

When it happens

Trigger: Running the memryx detector with memx_model_type set to a type with no post_process branch (e.g. a custom/other enum value), then detections arrive and process_output is called.

Common situations: Custom model zip with a model type enum the plugin does not post-process; config copy/pasted from another detector's model metadata; newer model type not yet supported by the installed Frigate version.

Related errors

AI-assisted analysis of blakeblackshear/frigate@ca18b8dc13 (2026-08-27). Data as JSON: /api/errors/44e24ae88cce3123. Report an issue: GitHub.