Distil-Bert-Base-Uncased-Hf: Optimized for Qualcomm Devices

DistilBERT is a lightweight BERT model designed for efficient self-supervised learning of language representations. It can be used for masked language modeling and as a backbone for various NLP tasks.

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
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a16	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 Distil-Bert-Base-Uncased-Hf 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 Distil-Bert-Base-Uncased-Hf on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.text_generation

Model Stats:

• Model checkpoint: distil_bert_base_uncased_hf
• Input resolution: 1x384
• Number of parameters: 11.3M
• Model size (float): 43.3 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Distil-Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	8.807 ms	0 - 525 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® X2 Elite	9.069 ms	178 - 178 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® X Elite	18.653 ms	178 - 178 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® 8 Gen 3 Mobile	14.927 ms	0 - 550 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	float	Qualcomm® QCS8550 (Proxy)	19.695 ms	0 - 197 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	float	Qualcomm® QCS9075	22.55 ms	0 - 3 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	10.459 ms	0 - 481 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	5.028 ms	0 - 306 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® X2 Elite	5.052 ms	111 - 111 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® X Elite	12.041 ms	112 - 112 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	8.406 ms	0 - 398 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCS6490	1346.808 ms	188 - 278 MB	CPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	11.443 ms	0 - 116 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCS9075	12.4 ms	0 - 3 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCM6690	716.189 ms	350 - 363 MB	CPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	6.311 ms	0 - 300 MB	NPU
Distil-Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	702.226 ms	355 - 367 MB	CPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	6.641 ms	0 - 383 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® X2 Elite	7.39 ms	1 - 1 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® X Elite	14.516 ms	0 - 0 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	11.751 ms	0 - 424 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	50.149 ms	0 - 384 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	15.446 ms	0 - 2 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® SA8775P	18.8 ms	0 - 385 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS9075	19.163 ms	0 - 2 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	36.782 ms	0 - 409 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® SA7255P	50.149 ms	0 - 384 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® SA8295P	23.144 ms	0 - 363 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	8.198 ms	0 - 383 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.969 ms	0 - 293 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® X2 Elite	3.589 ms	1 - 1 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® X Elite	7.894 ms	0 - 0 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	5.228 ms	0 - 357 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	17.894 ms	0 - 290 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	7.39 ms	0 - 300 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® SA8775P	7.59 ms	0 - 290 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® QCS9075	9.269 ms	0 - 2 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® SA7255P	17.894 ms	0 - 290 MB	NPU
Distil-Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	4.115 ms	0 - 290 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	6.923 ms	0 - 392 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	11.971 ms	0 - 437 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS8275 (Proxy)	50.294 ms	0 - 398 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS8550 (Proxy)	15.178 ms	0 - 3 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® SA8775P	19.042 ms	0 - 395 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS9075	19.533 ms	0 - 176 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS8450 (Proxy)	36.857 ms	0 - 408 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® SA7255P	50.294 ms	0 - 398 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® SA8295P	23.576 ms	0 - 368 MB	NPU
Distil-Bert-Base-Uncased-Hf	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	8.635 ms	0 - 388 MB	NPU

License

• The license for the original implementation of Distil-Bert-Base-Uncased-Hf can be found here.

References

• DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter

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.