---
title: DETR-ResNet101
canonical_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/DETR-ResNet101"
md_url: "https://www.modelscope.cn/models/Qualcomm-AI-Hub/DETR-ResNet101.md"
repository: Qualcomm-AI-Hub/DETR-ResNet101
last_updated: 2026-10-16
license: apache-2.0
downloads: 0
stars: 0
tags:
  - bu_auto
  - android
---

# DETR-ResNet101

> DETR-ResNet101 - Qualcomm-AI-Hub 在 ModelScope 开源的模型。DETR-ResNet101: Optimized for Qualcomm Devices

Qualcomm-AI-Hub/DETR-ResNet101 是 ModelScope 魔搭社区上的机器学习模型，采用 apache-2.0 许可。

- **Repository**: Qualcomm-AI-Hub/DETR-ResNet101
- **License**: apache-2.0
- **Tags**: bu_auto, android
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-10-16

Source: https://www.modelscope.cn/models/Qualcomm-AI-Hub/DETR-ResNet101

---

![](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet101/web-assets/model_demo.png)

# DETR-ResNet101: Optimized for Qualcomm Devices

DETR is a machine learning model that can detect objects (trained on COCO dataset).

This is based on the implementation of DETR-ResNet101 found [here](https://github.com/facebookresearch/detr).
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/detr_resnet101) 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 |
|---|---|---|---|---|
| ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet101/releases/v0.62.1/detr_resnet101-onnx-float.zip)
| ONNX | w8a16_mixed_int16 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet101/releases/v0.62.1/detr_resnet101-onnx-w8a16_mixed_int16.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet101/releases/v0.62.1/detr_resnet101-qnn_dlc-float.zip)
| QNN_DLC | w8a16_mixed_int16 | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet101/releases/v0.62.1/detr_resnet101-qnn_dlc-w8a16_mixed_int16.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://china-qaihub-public-assets.s3.cn-northwest-1.amazonaws.com.cn/qai-hub-models/models/detr_resnet101/releases/v0.62.1/detr_resnet101-tflite-float.zip)

For more device-specific assets and performance metrics, visit **[DETR-ResNet101 on Qualcomm® AI Hub](https://aihub.qualcomm.cn/models/detr_resnet101)**.


### 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/detr_resnet101) 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 [DETR-ResNet101 on GitHub](https://github.com/qualcomm/ai-hub-models-cn/blob/main/src/qai_hub_models/models/detr_resnet101) for usage instructions.

## Model Details

**Model Type:** Model_use_case.object_detection

**Model Stats:**
- Input resolution: 480x480
- Model checkpoint: ResNet101
- Model size (float): 230 MB
- Number of parameters: 60.3M

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| DETR-ResNet101 | ONNX | float | Snapdragon® X2 Elite | 12.64 ms | 5 - 5 MB | NPU
| DETR-ResNet101 | ONNX | float | Snapdragon® X Elite | 24.808 ms | 115 - 115 MB | NPU
| DETR-ResNet101 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 18.38 ms | 0 - 453 MB | NPU
| DETR-ResNet101 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 49.626 ms | 1 - 362 MB | NPU
| DETR-ResNet101 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 42.973 ms | 5 - 13 MB | NPU
| DETR-ResNet101 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 24.775 ms | 0 - 126 MB | NPU
| DETR-ResNet101 | ONNX | float | Qualcomm® QCS8450 | 49.626 ms | 1 - 362 MB | NPU
| DETR-ResNet101 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 39.121 ms | 5 - 12 MB | NPU
| DETR-ResNet101 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 24.808 ms | 115 - 115 MB | NPU
| DETR-ResNet101 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 14.546 ms | 3 - 330 MB | NPU
| DETR-ResNet101 | ONNX | float | Snapdragon® 8 Elite Mobile | 14.546 ms | 3 - 330 MB | NPU
| DETR-ResNet101 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.804 ms | 3 - 342 MB | NPU
| DETR-ResNet101 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 48.287 ms | 13 - 464 MB | NPU
| DETR-ResNet101 | ONNX | w8a16_mixed_int16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 56.86 ms | 0 - 97 MB | NPU
| DETR-ResNet101 | ONNX | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-8750 | 42.265 ms | 13 - 314 MB | NPU
| DETR-ResNet101 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite Mobile | 42.265 ms | 13 - 314 MB | NPU
| DETR-ResNet101 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 34.07 ms | 13 - 362 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Snapdragon® X2 Elite | 12.699 ms | 5 - 5 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Snapdragon® X Elite | 27.048 ms | 5 - 5 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 19.766 ms | 0 - 435 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 54.73 ms | 3 - 342 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 44.211 ms | 5 - 12 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 26.587 ms | 5 - 7 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA8775P | 41.148 ms | 1 - 318 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA8650P | 41.148 ms | 1 - 318 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA8255P | 41.148 ms | 1 - 318 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® QCS8450 | 54.73 ms | 3 - 342 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 41.517 ms | 5 - 11 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 27.048 ms | 5 - 5 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 15.271 ms | 0 - 323 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA7255P | 138.019 ms | 0 - 317 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Qualcomm® SA8295P | 43.204 ms | 0 - 233 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 15.271 ms | 0 - 323 MB | NPU
| DETR-ResNet101 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.052 ms | 5 - 345 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X2 Elite | 14.161 ms | 2 - 2 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X Elite | 26.067 ms | 2 - 2 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 19.996 ms | 2 - 501 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ IQ-8275 | 25.76 ms | 2 - 7 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 25.163 ms | 2 - 381 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8775P | 25.919 ms | 0 - 364 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8650P | 25.919 ms | 0 - 364 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8255P | 25.919 ms | 0 - 364 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ IQ-9075 | 26.292 ms | 2 - 6 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ IQ-X7181 | 26.067 ms | 2 - 2 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-6690 | 426.074 ms | 3 - 470 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-7790 | 38.735 ms | 2 - 434 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® Dragonwing™ Q-8750 | 15.504 ms | 0 - 349 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA7255P | 65.653 ms | 2 - 366 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite Mobile | 15.504 ms | 0 - 349 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 11.846 ms | 2 - 446 MB | NPU
| DETR-ResNet101 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 38.735 ms | 2 - 434 MB | NPU
| DETR-ResNet101 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 19.822 ms | 0 - 497 MB | NPU
| DETR-ResNet101 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 55.584 ms | 0 - 387 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 44.201 ms | 0 - 126 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 26.458 ms | 0 - 3 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® SA8775P | 41.253 ms | 0 - 367 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® SA8650P | 41.253 ms | 0 - 367 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® SA8255P | 41.253 ms | 0 - 367 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® QCS8450 | 55.584 ms | 0 - 387 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 41.399 ms | 0 - 125 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 15.296 ms | 0 - 375 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® SA7255P | 138.144 ms | 0 - 367 MB | NPU
| DETR-ResNet101 | TFLITE | float | Qualcomm® SA8295P | 43.292 ms | 0 - 275 MB | NPU
| DETR-ResNet101 | TFLITE | float | Snapdragon® 8 Elite Mobile | 15.296 ms | 0 - 375 MB | NPU
| DETR-ResNet101 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 11.095 ms | 0 - 392 MB | NPU

## License
* The license for the original implementation of DETR-ResNet101 can be found
  [here](https://github.com/facebookresearch/detr/blob/main/LICENSE).

## References
* [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872)
* [Source Model Implementation](https://github.com/facebookresearch/detr)

## Community
* For questions or feedback, chat with our engineers in [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.cn).