Model Zoo for Edge AI Applications
MERA developers get a head start with our Model Zoo, pre-trained AI inference models optimized for EdgeCortix AI accelerator chips or FPGAs with DNA IP. Code drops into MERA, ready to run or modify. Applications include classification, object detection, segmentation, pose estimation, and more.
Featured AI Inference Models
Filter by:
Clear Filters
No models are available for the selected combination. Please 'Clear Filters' to search again.
Model
ResNet18-v1.5Framework
PyTorchApplication
ClassificationInput Resolution
224x224Calibration Data
Real-DataGet the Model
Model
ResNet50-v1.5Framework
PyTorchApplication
ClassificationInput Resolution
224x224Calibration Data
Real-DataGet the Model
Model
YoloV3Framework
TFLiteApplication
Object-DetectionInput Resolution
416x416Calibration Data
Real-DataGet the Model
Model
Yolov5sFramework
TFLiteApplication
Object-DetectionInput Resolution
448x448Calibration Data
Real-DataGet the Model
Model
YoloV5mFramework
TFLiteApplication
Object-DetectionInput Resolution
640x640Calibration Data
Real-DataGet the Model
Model
SFA3DFramework
PyTorchApplication
3D-LiDAR-Object-DetectionInput Resolution
608x608Calibration Data
Real-DataGet the Model
Model
EfficientNet-Lite-0Framework
TFLiteApplication
ClassificationInput Resolution
240x240Calibration Data
Real-DataGet the Model
Model
EfficientNet-Lite-2Framework
TFLiteApplication
ClassificationInput Resolution
260x260Calibration Data
Real-DataGet the Model
Model
EfficientNet-Lite-3Framework
TFLiteApplication
ClassificationInput Resolution
280x280Calibration Data
Real-DataGet the Model
Model
EfficientNet-Lite-4Framework
TFLiteApplication
ClassificationInput Resolution
300x300Calibration Data
Real-DataGet the Model
Model
EfficientNetV2-B0Framework
TFLiteApplication
ClassificationInput Resolution
224x224Calibration Data
Random-DataGet the Model
Model
EfficientNetV2-B1Framework
TFLiteApplication
ClassificationInput Resolution
224x224Calibration Data
Random-DataGet the Model
Model
EfficientNetV2-B2Framework
TFLiteApplication
ClassificationInput Resolution
224x224Calibration Data
Random-DataGet the Model
Model
EfficientNetV2-B3Framework
TFLiteApplication
ClassificationInput Resolution
224x224Calibration Data
Random-DataGet the Model
Model
EfficientNetV2-sFramework
TFLiteApplication
ClassificationInput Resolution
224x224Calibration Data
Random-DataGet the Model
Model
MonoDepthFramework
PyTorchApplication
Monocular-Depth-EstimationInput Resolution
384x288Calibration Data
Real-DataGet the Model
Model
U-NetFramework
TFLiteApplication
SegmentationInput Resolution
128x128Calibration Data
Real-DataGet the Model
Model
MoveNet-ThunderFramework
TFLiteApplication
Pose-EstimationInput Resolution
256x256Calibration Data
Real-DataGet the Model
Model
YoloV4-TinyFramework
TFLiteApplication
Object-DetectionInput Resolution
640x640Calibration Data
Real-DataGet the Model
Model
DeepLabEdgeTPU-mFramework
TFLiteApplication
SegmentationInput Resolution
512x512Calibration Data
Real-DataGet the Model
Model
DeepLabEdgeTPU-sFramework
TFLiteApplication
SegmentationInput Resolution
512x512Calibration Data
Real-DataGet the Model
Model
MoveNet-LightingFramework
TFLiteApplication
Pose-EstimationInput Resolution
192x192Calibration Data
Real-DataGet the Model
Model
MobileNetV2-SSDFramework
PyTorchApplication
Object-DetectionInput Resolution
640x480Calibration Data
Real-DataGet the Model
Model
DeepLabEdgeTPU-xsFramework
TFLiteApplication
SegmentationInput Resolution
512x512Calibration Data
Real-DataGet the Model
Model
GladNetFramework
TFLiteApplication
Low-Light-EnhancementInput Resolution
640x480Calibration Data
Real-DataGet the Model
Model
SR-Mobile-Quantization (ABPN) Framework
TFLiteApplication
Super-ResolutionInput Resolution
640x360 to HDCalibration Data
Real-DataGet the Model
Model
YoloV7-QuantizerFramework
MERAApplication
Object-DetectionInput Resolution
640x640Calibration Data
Real-DataGet the Model
Model
YoloV4Framework
TFLiteApplication
Object-DetectionInput Resolution
416x416Calibration Data
Real-DataGet the Model
Model
SCI-QuantizerFramework
TFLiteApplication
Low-Light-EnhancementInput Resolution
1280x720Calibration Data
Real-DataGet the Model
SAKURA-I PCIe Low Profile Development Card
EdgeCortix SAKURA-I is available on a PCIe Low Profile development card, ready to drop into a host for software development and AI model inference tasks.
