Depth-Anything-V3: Optimized for Qualcomm Devices

Depth Anything 3 (DA3), a model that predicts spatially consistent geometry from arbitrary visual inputs, with or without known camera poses.

This is based on the implementation of Depth-Anything-V3 found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime	Precision	Chipset	SDK Versions	Download
ONNX	float	Universal	QAIRT 2.42, ONNX Runtime 1.25.0	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit Depth-Anything-V3 on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

• Custom weights (e.g., fine-tuned checkpoints)
• Custom input shapes
• Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for Depth-Anything-V3 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.depth_estimation

Model Stats:

• Model checkpoint: da3-small
• Input resolution: 518x518
• Number of parameters: 24.7M
• Model size (float): 94.3 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Depth-Anything-V3	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	29.608 ms	5 - 579 MB	NPU
Depth-Anything-V3	ONNX	float	Snapdragon® X2 Elite	28.021 ms	209 - 209 MB	NPU
Depth-Anything-V3	ONNX	float	Snapdragon® X Elite	68.397 ms	146 - 146 MB	NPU
Depth-Anything-V3	ONNX	float	Snapdragon® 8 Gen 3 Mobile	49.107 ms	5 - 736 MB	NPU
Depth-Anything-V3	ONNX	float	Qualcomm® QCS8550 (Proxy)	67.253 ms	4 - 123 MB	NPU
Depth-Anything-V3	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	35.85 ms	2 - 567 MB	NPU
Depth-Anything-V3	ONNX	float	Qualcomm® QCS9075	80.107 ms	4 - 49 MB	NPU
Depth-Anything-V3	ONNX	float	Qualcomm® QCS8750	35.85 ms	2 - 567 MB	NPU
Depth-Anything-V3	ONNX	float	Qualcomm® QCS7181	68.397 ms	146 - 146 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	33.907 ms	3 - 517 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Snapdragon® X2 Elite	35.09 ms	3 - 3 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Snapdragon® X Elite	78.86 ms	3 - 3 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	61.436 ms	2 - 647 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® QCS8275	187.343 ms	2 - 493 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	78.836 ms	3 - 5 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® SA8775P	85.33 ms	1 - 491 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® SA8650P	85.33 ms	1 - 491 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® SA8255P	85.33 ms	1 - 491 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	152.219 ms	2 - 677 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® SA7255P	187.343 ms	2 - 493 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® SA8295P	118.932 ms	0 - 504 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	42.862 ms	2 - 504 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® QCS9075	98.051 ms	3 - 9 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® QCS8750	42.862 ms	2 - 504 MB	NPU
Depth-Anything-V3	QNN_DLC	float	Qualcomm® QCS7181	78.86 ms	3 - 3 MB	NPU
Depth-Anything-V3	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	31.303 ms	1 - 485 MB	NPU
Depth-Anything-V3	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	54.49 ms	1 - 600 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® QCS8275	180.85 ms	1 - 467 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® QCS8550 (Proxy)	74.411 ms	0 - 19 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® SA8775P	81.174 ms	1 - 467 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® SA8650P	81.174 ms	1 - 467 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® SA8255P	81.174 ms	1 - 467 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® QCS8450 (Proxy)	119.406 ms	0 - 610 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® SA7255P	180.85 ms	1 - 467 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® SA8295P	103.526 ms	1 - 466 MB	NPU
Depth-Anything-V3	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	39.096 ms	1 - 475 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® QCS9075	95.329 ms	0 - 80 MB	NPU
Depth-Anything-V3	TFLITE	float	Qualcomm® QCS8750	39.096 ms	1 - 475 MB	NPU

License

• The license for the original implementation of Depth-Anything-V3 can be found here.

References

• Depth Anything 3: Recovering the visual space from any views
• Source Model Implementation

Community

• Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
• For questions or feedback please reach out to us.