--- title: SwinV2-Base canonical_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/SwinV2-Base" md_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/SwinV2-Base.md" repository: Qualcomm-AI-Hub/SwinV2-Base last_updated: 2026-10-16 license: bsd-3-clause pipeline_tag: image-classification tasks: - image-classification downloads: 0 stars: 0 tags: - backbone - bu_auto - android --- # SwinV2-Base > SwinV2-Base - Qualcomm-AI-Hub 在 ModelScope 开源的模型。SwinV2-Base: Optimized for Qualcomm Devices Qualcomm-AI-Hub/SwinV2-Base 是 ModelScope 魔搭社区上的image-classification模型，采用 bsd-3-clause 许可。 - **Repository**: Qualcomm-AI-Hub/SwinV2-Base - **License**: bsd-3-clause - **Tasks**: image-classification - **Tags**: backbone, bu_auto, android - **Downloads**: 0 - **Stars**: 0 - **Last updated**: 2026-10-16 Source: https://www.modelscope.cn/models/Qualcomm-AI-Hub/SwinV2-Base --- ![](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/swinv2_base/web-assets/model_demo.png) # SwinV2-Base: Optimized for Qualcomm Devices SwinV2Base is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases. This is based on the implementation of SwinV2-Base found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py). This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/swinv2_base) 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.cn) 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 | |---|---|---|---|---| | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/swinv2_base/releases/v0.62.1/swinv2_base-qnn_dlc-float.zip) | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/swinv2_base/releases/v0.62.1/swinv2_base-qnn_dlc-w8a16.zip) | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/swinv2_base/releases/v0.62.1/swinv2_base-tflite-float.zip) For more device-specific assets and performance metrics, visit **[SwinV2-Base on Qualcomm® AI Hub](https://aihub.qualcomm.cn/models/swinv2_base)**. ### Option 2: Export with Custom Configurations Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/swinv2_base) 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 [SwinV2-Base on GitHub](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/swinv2_base) for usage instructions. ## Model Details **Model Type:** Model_use_case.image_classification **Model Stats:** - Input resolution: 256x256 - Model checkpoint: Imagenet - Model size (float): 339 MB - Model size (w8a16): 90.2 MB - Number of parameters: 88.8M ## Performance Summary | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |---|---|---|---|---|---|--- | SwinV2-Base | QNN_DLC | float | Snapdragon® X2 Elite | 12.502 ms | 1 - 1 MB | NPU | SwinV2-Base | QNN_DLC | float | Snapdragon® X Elite | 28.992 ms | 1 - 1 MB | NPU | SwinV2-Base | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 19.584 ms | 0 - 545 MB | NPU | SwinV2-Base | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 41.306 ms | 0 - 532 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 30.155 ms | 1 - 4 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 27.793 ms | 1 - 3 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® SA8775P | 31.465 ms | 1 - 388 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® SA8650P | 31.465 ms | 1 - 388 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® SA8255P | 31.465 ms | 1 - 388 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® QCS8450 | 41.306 ms | 0 - 532 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 31.753 ms | 1 - 3 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 28.992 ms | 1 - 1 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 14.685 ms | 1 - 389 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® SA7255P | 74.058 ms | 1 - 394 MB | NPU | SwinV2-Base | QNN_DLC | float | Qualcomm® SA8295P | 37.76 ms | 1 - 378 MB | NPU | SwinV2-Base | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 14.685 ms | 1 - 389 MB | NPU | SwinV2-Base | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.588 ms | 0 - 426 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 12.179 ms | 0 - 0 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Snapdragon® X Elite | 30.75 ms | 0 - 0 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 19.661 ms | 0 - 2034 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 25.972 ms | 0 - 3 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 29.407 ms | 0 - 3 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® SA8775P | 30.025 ms | 0 - 869 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® SA8650P | 30.025 ms | 0 - 869 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® SA8255P | 30.025 ms | 0 - 869 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 29.394 ms | 0 - 2 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 30.75 ms | 0 - 0 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 14.803 ms | 0 - 902 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Qualcomm® SA7255P | 52.507 ms | 0 - 910 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 14.803 ms | 0 - 902 MB | NPU | SwinV2-Base | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 11.466 ms | 0 - 953 MB | NPU | SwinV2-Base | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 20.09 ms | 0 - 2182 MB | NPU | SwinV2-Base | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 42.556 ms | 0 - 675 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 29.908 ms | 0 - 181 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 29.425 ms | 0 - 4 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® SA8775P | 32.539 ms | 0 - 878 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® SA8650P | 32.539 ms | 0 - 878 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® SA8255P | 32.539 ms | 0 - 878 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® QCS8450 | 42.556 ms | 0 - 675 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 33.009 ms | 0 - 181 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 14.968 ms | 0 - 894 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® SA7255P | 71.141 ms | 0 - 884 MB | NPU | SwinV2-Base | TFLITE | float | Qualcomm® SA8295P | 40.597 ms | 0 - 875 MB | NPU | SwinV2-Base | TFLITE | float | Snapdragon® 8 Elite Mobile | 14.968 ms | 0 - 894 MB | NPU | SwinV2-Base | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.375 ms | 0 - 939 MB | NPU ## License * The license for the original implementation of SwinV2-Base can be found [here](https://github.com/pytorch/vision/blob/main/LICENSE). ## References * [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) * [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py) ## Community * For questions or feedback, chat with our engineers in [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.cn).