FastSam-X: Optimized for Qualcomm Devices

The Fast Segment Anything Model (FastSAM) is a novel, real-time CNN-based solution for the Segment Anything task. This task is designed to segment any object within an image based on various possible user interaction prompts. The model performs competitively despite significantly reduced computation, making it a practical choice for a variety of vision tasks.

This is based on the implementation of FastSam-X 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 FastSam-X 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 FastSam-X on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

• Model checkpoint: fastsam-x.pt
• Inference latency: RealTime
• Input resolution: 640x640
• Number of parameters: 72.2M
• Model size (float): 276 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
FastSam-X	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	19.417 ms	17 - 260 MB	NPU
FastSam-X	ONNX	float	Snapdragon® X2 Elite	24.056 ms	207 - 207 MB	NPU
FastSam-X	ONNX	float	Snapdragon® X Elite	46.402 ms	144 - 144 MB	NPU
FastSam-X	ONNX	float	Snapdragon® 8 Gen 3 Mobile	35.829 ms	13 - 332 MB	NPU
FastSam-X	ONNX	float	Qualcomm® QCS8550 (Proxy)	46.859 ms	0 - 412 MB	NPU
FastSam-X	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	27.549 ms	13 - 247 MB	NPU
FastSam-X	ONNX	float	Qualcomm® QCS9075	74.018 ms	13 - 58 MB	NPU
FastSam-X	ONNX	float	Qualcomm® QCS8750	27.549 ms	13 - 247 MB	NPU
FastSam-X	ONNX	float	Qualcomm® QCS7181	46.402 ms	144 - 144 MB	NPU
FastSam-X	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	17.46 ms	5 - 236 MB	NPU
FastSam-X	QNN_DLC	float	Snapdragon® X2 Elite	23.186 ms	5 - 5 MB	NPU
FastSam-X	QNN_DLC	float	Snapdragon® X Elite	44.05 ms	5 - 5 MB	NPU
FastSam-X	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	32.779 ms	2 - 299 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® QCS8275	279.927 ms	0 - 215 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	42.926 ms	5 - 7 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® SA8775P	68.473 ms	0 - 214 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® SA8650P	68.473 ms	0 - 214 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® SA8255P	68.473 ms	0 - 214 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	92.634 ms	0 - 381 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® SA7255P	279.927 ms	0 - 215 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® SA8295P	77.648 ms	0 - 298 MB	NPU
FastSam-X	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	25.358 ms	0 - 223 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® QCS9075	71.002 ms	5 - 15 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® QCS8750	25.358 ms	0 - 223 MB	NPU
FastSam-X	QNN_DLC	float	Qualcomm® QCS7181	44.05 ms	5 - 5 MB	NPU
FastSam-X	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	17.225 ms	4 - 262 MB	NPU
FastSam-X	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	32.87 ms	4 - 420 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® QCS8275	279.447 ms	5 - 256 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® QCS8550 (Proxy)	42.441 ms	4 - 7 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® SA8775P	67.934 ms	5 - 257 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® SA8650P	67.934 ms	5 - 257 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® SA8255P	67.934 ms	5 - 257 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® QCS8450 (Proxy)	92.492 ms	5 - 507 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® SA7255P	279.447 ms	5 - 256 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® SA8295P	77.169 ms	0 - 330 MB	NPU
FastSam-X	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	25.42 ms	3 - 261 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® QCS9075	69.975 ms	4 - 158 MB	NPU
FastSam-X	TFLITE	float	Qualcomm® QCS8750	25.42 ms	3 - 261 MB	NPU

License

• The license for the original implementation of FastSam-X can be found here.

References

• Fast Segment Anything
• 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.