FastSam-S: 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-S 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 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	Download

For more device-specific assets and performance metrics, visit FastSam-S 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-S on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

• Model checkpoint: fastsam-s.pt
• Inference latency: RealTime
• Input resolution: 640x640
• Number of parameters: 11.8M
• Model size (float): 45.1 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
FastSam-S	ONNX	float	Snapdragon® X Elite	8.597 ms	19 - 19 MB	NPU
FastSam-S	ONNX	float	Snapdragon® 8 Gen 3 Mobile	5.956 ms	14 - 280 MB	NPU
FastSam-S	ONNX	float	Qualcomm® QCS8550 (Proxy)	8.023 ms	0 - 26 MB	NPU
FastSam-S	ONNX	float	Qualcomm® QCS9075	13.16 ms	12 - 15 MB	NPU
FastSam-S	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	4.481 ms	11 - 236 MB	NPU
FastSam-S	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	3.367 ms	0 - 211 MB	NPU
FastSam-S	ONNX	float	Snapdragon® X2 Elite	4.337 ms	20 - 20 MB	NPU
FastSam-S	QNN_DLC	float	Snapdragon® X Elite	8.042 ms	5 - 5 MB	NPU
FastSam-S	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	5.546 ms	5 - 214 MB	NPU
FastSam-S	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	38.7 ms	1 - 183 MB	NPU
FastSam-S	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	7.465 ms	5 - 6 MB	NPU
FastSam-S	QNN_DLC	float	Qualcomm® SA8775P	11.283 ms	0 - 188 MB	NPU
FastSam-S	QNN_DLC	float	Qualcomm® QCS9075	10.878 ms	5 - 15 MB	NPU
FastSam-S	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	15.096 ms	4 - 211 MB	NPU
FastSam-S	QNN_DLC	float	Qualcomm® SA7255P	38.7 ms	1 - 183 MB	NPU
FastSam-S	QNN_DLC	float	Qualcomm® SA8295P	13.897 ms	0 - 176 MB	NPU
FastSam-S	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	4.31 ms	0 - 185 MB	NPU
FastSam-S	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	3.127 ms	5 - 192 MB	NPU
FastSam-S	QNN_DLC	float	Snapdragon® X2 Elite	4.477 ms	5 - 5 MB	NPU
FastSam-S	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	5.092 ms	3 - 121 MB	NPU
FastSam-S	TFLITE	float	Qualcomm® QCS8275 (Proxy)	37.683 ms	4 - 82 MB	NPU
FastSam-S	TFLITE	float	Qualcomm® QCS8550 (Proxy)	6.838 ms	4 - 6 MB	NPU
FastSam-S	TFLITE	float	Qualcomm® SA8775P	10.635 ms	4 - 85 MB	NPU
FastSam-S	TFLITE	float	Qualcomm® QCS9075	10.592 ms	4 - 39 MB	NPU
FastSam-S	TFLITE	float	Qualcomm® QCS8450 (Proxy)	14.034 ms	4 - 234 MB	NPU
FastSam-S	TFLITE	float	Qualcomm® SA7255P	37.683 ms	4 - 82 MB	NPU
FastSam-S	TFLITE	float	Qualcomm® SA8295P	13.114 ms	4 - 195 MB	NPU
FastSam-S	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	3.869 ms	3 - 97 MB	NPU
FastSam-S	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	2.893 ms	0 - 201 MB	NPU

License

• The license for the original implementation of FastSam-S 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.