--- library_name: pytorch license: other tags: - backbone - android pipeline_tag: text-generation --- ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/albert_base_v2_hf/web-assets/model_demo.png) # Albert-Base-V2-Hf: Optimized for Qualcomm Devices ALBERT 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 Albert-Base-V2-Hf found [here](https://github.com/google-research/albert). This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/albert_base_v2_hf) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary). Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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 2.37, ONNX Runtime 1.23.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/albert_base_v2_hf/releases/v0.46.0/albert_base_v2_hf-onnx-float.zip) | QNN_DLC | float | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/albert_base_v2_hf/releases/v0.46.0/albert_base_v2_hf-qnn_dlc-float.zip) | QNN_DLC | w8a16 | Universal | QAIRT 2.42 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/albert_base_v2_hf/releases/v0.46.0/albert_base_v2_hf-qnn_dlc-w8a16.zip) | TFLITE | float | Universal | QAIRT 2.42, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/albert_base_v2_hf/releases/v0.46.0/albert_base_v2_hf-tflite-float.zip) For more device-specific assets and performance metrics, visit **[Albert-Base-V2-Hf on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/albert_base_v2_hf)**. ### Option 2: Export with Custom Configurations Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/albert_base_v2_hf) 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 [Albert-Base-V2-Hf on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/albert_base_v2_hf) for usage instructions. ## Model Details **Model Type:** Model_use_case.text_generation **Model Stats:** - Model checkpoint: albert/albert-base-v2 - Input resolution: 1x384 - Number of parameters: 11.8M - Model size (float): 43.9 MB ## Performance Summary | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |---|---|---|---|---|---|--- | Albert-Base-V2-Hf | ONNX | float | Snapdragon® X Elite | 28.987 ms | 33 - 33 MB | NPU | Albert-Base-V2-Hf | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 22.278 ms | 0 - 385 MB | NPU | Albert-Base-V2-Hf | ONNX | float | Qualcomm® QCS8550 (Proxy) | 30.096 ms | 0 - 317 MB | NPU | Albert-Base-V2-Hf | ONNX | float | Qualcomm® QCS9075 | 32.316 ms | 0 - 3 MB | NPU | Albert-Base-V2-Hf | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 17.079 ms | 0 - 326 MB | NPU | Albert-Base-V2-Hf | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.437 ms | 0 - 335 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Snapdragon® X Elite | 22.358 ms | 0 - 0 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 17.408 ms | 0 - 375 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 74.867 ms | 0 - 317 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 23.222 ms | 0 - 2 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Qualcomm® SA8775P | 27.658 ms | 0 - 316 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Qualcomm® QCS9075 | 26.841 ms | 0 - 2 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 35.81 ms | 0 - 421 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Qualcomm® SA7255P | 74.867 ms | 0 - 317 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Qualcomm® SA8295P | 33.432 ms | 0 - 376 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 11.917 ms | 0 - 386 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 9.823 ms | 0 - 391 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Snapdragon® X Elite | 13.997 ms | 0 - 0 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 9.558 ms | 0 - 295 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 29.59 ms | 0 - 257 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 13.542 ms | 0 - 236 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Qualcomm® SA8775P | 13.392 ms | 0 - 249 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 15.845 ms | 0 - 2 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Qualcomm® SA7255P | 29.59 ms | 0 - 257 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 7.916 ms | 0 - 248 MB | NPU | Albert-Base-V2-Hf | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 5.495 ms | 0 - 265 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 17.443 ms | 0 - 387 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 74.61 ms | 0 - 321 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 22.268 ms | 0 - 343 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Qualcomm® SA8775P | 27.317 ms | 0 - 321 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Qualcomm® QCS9075 | 27.14 ms | 0 - 33 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 40.475 ms | 0 - 419 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Qualcomm® SA7255P | 74.61 ms | 0 - 321 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Qualcomm® SA8295P | 34.418 ms | 0 - 377 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 12.431 ms | 0 - 391 MB | NPU | Albert-Base-V2-Hf | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 10.065 ms | 0 - 388 MB | NPU ## License * The license for the original implementation of Albert-Base-V2-Hf can be found [here](https://github.com/google-research/albert/blob/master/LICENSE). ## References * [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942) * [Source Model Implementation](https://github.com/google-research/albert) ## Community * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI. * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).