Intel® Arc™ GPU for the Edge
Supercharge Edge AI, Graphics, and Media Processing with Intel Arc Graphics discrete GPU.
Boost Performance of Edge Workloads
Graphics
For immersive visual experiences.
AI Inferencing
Deploy advanced AI workloads at the edge with specialized AI engines.
Media Processing & Delivery
Enhance video production, media transcoding, and streaming.
What You Can Do With Intel Arc GPU
Deliver exceptional display density, resolution, and performance for demanding multi-monitor video wall applications with Matrox LUMA Pro Series graphics cards, powered by the Intel® Arc™ GPU.
The ADLINK automated optical inspection (AOI) solution is built with an Intel CPU, Intel GPUs, and ADLINK AI models that were optimized by OpenVINO toolkit. The solution provides highly accurate visual inference at scale to reduce overhead costs of quality assurance in manufacturing.
A370M Performance Compared to NVIDIA® T1000
Up to
2.4x
Higher performance at AI Inference using Resnet50, Batch Size 32. 1 2
Up to
2.28x
Higher performance at decoding 2 streams of H.264 1080p30 Video. 1 2
Up to
2.4x
Higher performance at AI Inference using Resnet50, Batch Size 32. 1 2
Up to
2.28x
Higher performance at decoding 2 streams of H.264 1080p30 Video. 1 2
Accelerating AI Inference for Developers
With the OpenVINO toolkit's optimization techniques, such as model quantization, layer fusion, and hardware-level optimizations, you can significantly enhance the efficiency of neural network inference. When deployed on a discrete GPU, these optimized models can leverage the parallel processing capabilities of the GPU, resulting in faster inference speeds.
Maintaining cost-efficiency while achieving exceptional GPU performance is made possible with OpenVINO. The latest OpenVINO 2023.1 release makes generative AI more accessible for real world scenarios with added broader model support, reduced memory usage, and the introduction of additional compression techniques for large language models (LLMs). Plus developers now have more flexibility working with their framework of choice. Download today and start creating AI solutions.
Product and Performance Information
See backup for workloads and configurations. Results may vary.
Test by Intel as of 31/08/23. 1-node, 1x 13th Gen Intel(R) Core(TM) i9-13900K, 24 cores, HT On, Turbo On, NUMA 1, Integrated Accelerators Available [used]: DLB 0 [0], DSA 0 [0], IAA 0 [0], QAT 0 [0], Total Memory 32GB (2x16GB DDR5 4800 MT/s [4800 MT/s]), BIOS A.C0, microcode 0x119, 1x Ethernet Controller I225-V, 1x 931.5G WDC WDS100T2B0B-00YS70, 1x 1.8T WDC WDS200T2B0B-00YS70, 1x 28.8G DataTraveler 2.0, Ubuntu 22.04.3 LTS, 6.2.0-26-generic, WORKLOAD+VERSION, COMPILER, LIBRARIES, OTHER_SW, score=?UNITS., NVRM version: NVIDIA UNIX x86_64 Kernel Module 535.86.10, NVIDIA Corporation TU117GL [T1000 8GB] (rev a1).
Test by Intel as of 31/08/23.1-node, 1x 12th Gen Intel(R) Core(TM) i7-12800HE, 14 cores, HT On, Turbo On, NUMA 1, Integrated Accelerators Available [used]: DLB 0 [0], DSA 0 [0], IAA 0 [0], QAT 0 [0], Total Memory 32GB (1x32GB DDR5 4800 MT/s [4800 MT/s]), BIOS 0.06.07, microcode 0x429, 1x Ethernet Connection (16) I219-LM, 1x 1.1T INTEL SSDSC2BB012T6, Ubuntu 22.04.3 LTS, 6.2.0-26-generic, WORKLOAD+VERSION, COMPILER, LIBRARIES, OTHER_SW, score=?UNITS. VGA compatible controller: Intel Corporation Device 5693 (rev 05).
Performance varies by use, configuration, and other factors. Learn more on the Performance Index site.