| framework | Backend | Device | Single | Half | Quant. | Results |
|---|---|---|---|---|---|---|
TensorFlow Lite | CPU | Intel Core i7-1355U | 1267 | 1118 | 804 | https://browser.geekbench.com/ai/v1/247577 |
ONNX | CPU | Intel Core i7-1355U | 1719 | 537 | 4011 | https://browser.geekbench.com/ai/v1/247584 |
OpenVINO | CPU | Intel Core i7-1355U | 1959 | 1876 | 3832 | https://browser.geekbench.com/ai/v1/247588 |
| Workload (TF Lite) | Accuracy | Score | Workload (ONNX) | Accuracy | Score | Workload | Accuracy | Score |
|---|---|---|---|---|---|---|---|---|
| Image Classification (SP) | 100% | 873 162.3 IPS | Image Classification (SP) | 100% | 1028 191.1 IPS | Image Classification (SP) | 100% | 1141 212.2 IPS |
| Image Classification (HP) | 100% | 865 160.9 IPS | Image Classification (HP) | 100% | 187 34.7 IPS | Image Classification (HP) | 100% | 1145 213.0 IPS |
| Image Classification (Q) | 99% | 546 101.8 IPS | Image Classification (Q) | 97% | 3315 618.6 IPS | Image Classification (Q) | 100% | 2573 478.4 IPS |
| Image Segmentation (SP) | 100% | 1136 18.4 IPS | Image Segmentation (SP) | 100% | 945 15.3 IPS | Image Segmentation (SP) | 100% | 1395 22.6 IPS |
| Image Segmentation (HP) | 100% | 1145 18.6 IPS | Image Segmentation (HP) | 100% | 291 4.71 IPS | Image Segmentation (HP) | 100% | 1443 23.4 IPS |
| Image Segmentation (Q) | 98% | 627 10.2 IPS | Image Segmentation (Q) | 99% | 2271 36.9 IPS | Image Segmentation (Q) | 99% | 2640 42.8 IPS |
| Pose Estimation (SP) | 100% | 1665 1.94 IPS | Pose Estimation (SP) | 100% | 2483 2.90 IPS | Pose Estimation (SP) | 100% | 2272 2.65 IPS |
| Pose Estimation (HP) | 100% | 1097 1.28 IPS | Pose Estimation (HP) | 100% | 1393 1.63 IPS | Pose Estimation (HP) | 100% | 1776 2.07 IPS |
| Pose Estimation (Q) | 96% | 1185 1.39 IPS | Pose Estimation (Q) | 94% | 7970 9.35 IPS | Pose Estimation (Q) | 96% | 5342 6.26 IPS |
| Object Detection (SP) | 100% | 892 70.7 IPS | Object Detection (SP) | 100% | 1124 89.2 IPS | Object Detection (SP) | 100% | 1125 89.2 IPS |
| Object Detection (HP) | 100% | 870 69.0 IPS | Object Detection (HP) | 100% | 194 15.4 IPS | Object Detection (HP) | 100% | 1128 89.4 IPS |
| Object Detection (Q) | 85% | 615 49.5 IPS | Object Detection (Q) | 86% | 3289 264.3 IPS | Object Detection (Q) | 88% | 2552 204.6 IPS |
| Face Detection (SP) | 100% | 1861 22.1 IPS | Face Detection (SP) | 100% | 2262 26.9 IPS | Face Detection (SP) | 100% | 3316 39.4 IPS |
| Face Detection (HP) | 100% | 1342 15.9 IPS | Face Detection (HP) | 100% | 279 3.32 IPS | Face Detection (HP) | 100% | 2781 33.0 IPS |
| Face Detection (Q) | 97% | 1515 18.1 IPS | Face Detection (Q) | 97% | 8407 100.2 IPS | Face Detection (Q) | 100% | 6399 76.0 IPS |
| Depth Estimation (SP) | 100% | 1663 12.8 IPS | Depth Estimation (SP) | 100% | 3168 24.4 IPS | Depth Estimation (SP) | 100% | 2723 21.0 IPS |
| Depth Estimation (HP) | 99% | 1589 12.2 IPS | Depth Estimation (HP) | 99% | 740 5.70 IPS | Depth Estimation (HP) | 99% | 2770 21.3 IPS |
| Depth Estimation (Q) | 63% | 1106 10.4 IPS | Depth Estimation (Q) | 78% | 7653 61.1 IPS | Depth Estimation (Q) | 89% | 6361 49.5 IPS |
| Style Transfer (SP) | 100% | 2077 2.67 IPS | Style Transfer (SP) | 100% | 5451 7.01 IPS | Style Transfer (SP) | 100% | 6421 8.25 IPS |
| Style Transfer (HP) | 100% | 1990 2.56 IPS | Style Transfer (HP) | 100% | 4078 5.24 IPS | Style Transfer (HP) | 100% | 6515 8.37 IPS |
| Style Transfer (Q) | 98% | 2812 3.63 IPS | Style Transfer (Q) | 98% | 10602 13.7 IPS | Style Transfer (Q) | 98% | 15385 19.8 IPS |
| Image Super-Resolution (SP) | 100% | 899 33.2 IPS | Image Super-Resolution (SP) | 100% | 1337 49.4 IPS | Image Super-Resolution (SP) | 100% | 1295 47.8 IPS |
| Image Super-Resolution (HP) | 100% | 907 33.5 IPS | Image Super-Resolution (HP) | 100% | 977 36.1 IPS | Image Super-Resolution (HP) | 100% | 1306 48.2 IPS |
| Image Super-Resolution (Q) | 97% | 714 26.5 IPS | Image Super-Resolution (Q) | 99% | 2829 104.7 IPS | Image Super-Resolution (Q) | 99% | 3217 119.1 IPS |
| Text Classification (SP) | 100% | 924 1.23 KIPS | Text Classification (SP) | 100% | 1085 1.45 KIPS | Text Classification (SP) | 100% | 1266 1.69 KIPS |
| Text Classification (HP) | 100% | 925 1.23 KIPS | Text Classification (HP) | 100% | 465 621.0 IPS | Text Classification (HP) | 100% | 1184 1.58 KIPS |
| Text Classification (Q) | 92% | 337 452.5 IPS | Text Classification (Q) | 97% | 1319 1.77 KIPS | Text Classification (Q) | 92% | 1764 2.37 KIPS |
| Machine Translation (SP) | 100% | 1362 23.5 IPS | Machine Translation (SP) | 100% | 1471 25.3 IPS | Machine Translation (SP) | 100% | 2158 37.2 IPS |
| Machine Translation (HP) | 100% | 1574 27.1 IPS | Machine Translation (HP) | 100% | 356 6.13 IPS | Machine Translation (HP) | 100% | 2105 36.3 IPS |
| Machine Translation (Q) | 58% | 401 9.41 IPS | Machine Translation (Q) | 65% | 2151 43.2 IPS | Machine Translation (Q) | 100% | 2078 35.8 IPS |
Install
# OpenVino wget https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB sudo apt-key add GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB echo "deb https://apt.repos.intel.com/openvino/2025 ubuntu24 main" | sudo tee /etc/apt/sources.list.d/intel-openvino-2025.list apt update apt-cache search openvino apt install openvino python3 /usr/share/openvino/samples/python/hello_query_device/hello_query_device.py # NPU driver git clone https://github.com/intel/linux-npu-driver.git cd linux-npu-driver/ apt install -y build-essential git git-lfs cmake python3 git submodule update --init --recursive cmake -B build -S . cmake --build build --parallel $(nproc) cmake --install build rmmod intel_vpu modprobe intel_vpu cmake -B build -S . cmake --install build/ --component fw-npu --prefix / #Geekbench mkdir Geekbench cd Geekbench wget https://cdn.geekbench.com/GeekbenchAI-1.3.0-Linux.tar.gz tar xvf GeekbenchAI-1.3.0-Linux.tar.gz cd GeekbenchAI-1.3.0-Linux/ |
Check Available frameworks
root@server4:~/Geekbench/GeekbenchAI-1.3.0-Linux# ./banff --ai-list Geekbench AI 1.3.0 : https://www.geekbench.com/ai/ Geekbench AI requires an active internet connection and automatically uploads benchmark results to the Geekbench Browser. Framework | Backend | Device 1 TensorFlow Lite | 1 CPU | 0 Intel Core i7-1355U 3 ONNX | 1 CPU | 0 Intel Core i7-1355U 4 OpenVINO | 1 CPU | 0 13th Gen Intel(R) Core(TM) i7-1355U 4 OpenVINO | 2 GPU | 1 Intel(R) Iris(R) Xe Graphics (iGPU) |
help
root@server5:/storage/apps/Geekbench/GeekbenchAI-1.3.0-Linux# ./banff --help Geekbench AI 1.3.0 : https://www.geekbench.com/ai/ Usage: ./banff [ options ] Options: -h, --help print this message AI Benchmark Options: --ai run the AI benchmark --ai-framework [ID] use AI framework ID --ai-backend [ID] use AI backend ID --ai-device [ID] use AI device ID --ai-list list available AI settings If no options are given, the default action is to run the inference benchmark. |
example run
root@server4:~/linux-npu-driver# modinfo intel_vpu filename: /lib/modules/6.8.0-57-generic/kernel/drivers/accel/ivpu/intel_vpu.ko.zst version: 1.0. license: GPL and additional rights description: Driver for Intel NPU (Neural Processing Unit) author: Intel Corporation firmware: intel/vpu/vpu_40xx_v0.0.bin firmware: intel/vpu/vpu_37xx_v0.0.bin srcversion: 853217D6461C2C5899F4F14 alias: pci:v00008086d0000643Esv*sd*bc*sc*i* alias: pci:v00008086d0000AD1Dsv*sd*bc*sc*i* alias: pci:v00008086d00007D1Dsv*sd*bc*sc*i* depends: retpoline: Y intree: Y name: intel_vpu |
root@server4:~/Geekbench/GeekbenchAI-1.3.0-Linux# ./banff --ai-framework 1 Geekbench AI 1.3.0 : https://www.geekbench.com/ai/ Geekbench AI requires an active internet connection and automatically uploads benchmark results to the Geekbench Browser. AI Information Framework TensorFlow Lite Backend CPU Device Intel Core i7-1355U System Information Operating System Ubuntu 24.04.2 LTS Model Default string Default string Motherboard Default string Default string BIOS American Megatrends International, LLC. 5.27 CPU Information Name Intel Core i7-1355U Topology 1 Processor, 10 Cores, 12 Threads Identifier GenuineIntel Family 6 Model 186 Stepping 3 Base Frequency 5.00 GHz Memory Information Size 94.1 GB Running Image Classification (SP) INFO: Initialized TensorFlow Lite runtime. INFO: Applying 1 TensorFlow Lite delegate(s) lazily. Running Image Classification (HP) INFO: Applying 1 TensorFlow Lite delegate(s) lazily. Running Image Classification (Q) Running Image Segmentation (SP) INFO: Applying 1 TensorFlow Lite delegate(s) lazily. Running Image Segmentation (HP) INFO: Applying 1 TensorFlow Lite delegate(s) lazily. Running Image Segmentation (Q) .. |