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

Bert 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 is based on the implementation of Bert-Base-Uncased-Hf 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
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 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 Bert-Base-Uncased-Hf on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.text_generation

Model Stats:

Model checkpoint: google-bert/bert-base-uncased
Input resolution: 1x384
Number of parameters: 110M
Model size (float): 418 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	13.772 ms	0 - 718 MB	NPU
Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® X2 Elite	14.671 ms	265 - 265 MB	NPU
Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® X Elite	31.087 ms	265 - 265 MB	NPU
Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® 8 Gen 3 Mobile	23.588 ms	0 - 748 MB	NPU
Bert-Base-Uncased-Hf	ONNX	float	Qualcomm® QCS8550 (Proxy)	31.393 ms	0 - 322 MB	NPU
Bert-Base-Uncased-Hf	ONNX	float	Qualcomm® QCS9075	35.899 ms	0 - 3 MB	NPU
Bert-Base-Uncased-Hf	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	16.783 ms	0 - 660 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	8.343 ms	0 - 410 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® X2 Elite	8.75 ms	154 - 154 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® X Elite	20.908 ms	154 - 154 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	14.824 ms	0 - 592 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCS6490	2296.756 ms	188 - 280 MB	CPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	19.907 ms	0 - 203 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCS9075	20.474 ms	0 - 3 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Qualcomm® QCM6690	1212.598 ms	198 - 212 MB	CPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	10.874 ms	0 - 417 MB	NPU
Bert-Base-Uncased-Hf	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	1189.815 ms	205 - 220 MB	CPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	9.467 ms	0 - 538 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® X2 Elite	10.614 ms	1 - 1 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® X Elite	22.514 ms	0 - 0 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	16.965 ms	0 - 584 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	81.5 ms	0 - 519 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	23.435 ms	0 - 287 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® SA8775P	28.695 ms	0 - 520 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS9075	28.729 ms	0 - 2 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	47.754 ms	0 - 567 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® SA7255P	81.5 ms	0 - 519 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Qualcomm® SA8295P	35.989 ms	0 - 501 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	11.864 ms	0 - 519 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	5.138 ms	0 - 414 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® X2 Elite	6.083 ms	1 - 1 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® X Elite	13.941 ms	0 - 0 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	9.14 ms	0 - 499 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	30.762 ms	0 - 408 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	13.264 ms	0 - 2 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® SA8775P	60.294 ms	0 - 408 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® QCS9075	15.641 ms	0 - 2 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Qualcomm® SA7255P	30.762 ms	0 - 408 MB	NPU
Bert-Base-Uncased-Hf	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	7.284 ms	0 - 409 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	9.746 ms	0 - 546 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	17.219 ms	0 - 597 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS8275 (Proxy)	81.834 ms	0 - 532 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS8550 (Proxy)	22.58 ms	0 - 3 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® SA8775P	28.883 ms	0 - 533 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS9075	28.838 ms	0 - 259 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® QCS8450 (Proxy)	32.538 ms	0 - 565 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® SA7255P	81.834 ms	0 - 532 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Qualcomm® SA8295P	35.491 ms	0 - 503 MB	NPU
Bert-Base-Uncased-Hf	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	12.214 ms	0 - 536 MB	NPU

License

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

References

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.

Downloads last month: -; Downloads are not tracked for this model. How to track

Paper for qualcomm/Bert-Base-Uncased-Hf

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Paper • 1810.04805 • Published Oct 11, 2018 • 26