Intel officially announced its Data Center GPU Flex series on Wednesday, but started shipping its Arctic Sound-M discrete processing units for data centers about a month ago. The new graphics boards are based on the company’s Arc Alchemist graphics processors, target a variety of data center applications, and will be available from various Intel partners once the products reach the required maturity.
Intel’s Data Center GPU Flex graphics card family includes two base products. The single-chip Flex Series 170 is based on one ACM-G10 GPU with up to 32 Xe cores (equivalent to up to 4,096 stream processors) and 16 GB of memory for workloads that demand maximum performance. Intended for loading. The dual-chip Flex Series 140 card features two ACM-G11 GPUs with 16 Xe cores and 12GB of memory and is targeted at high-density machines.
Intel Data Center GPU Flex Specifications
| Data Center GPU Flex 170 | Data Center GPU Flex 140 | |
| GPUs | ACM-G10 | 2x ACM-G11 |
| Xe core | 32 | 16 |
| execution unit | 512 | 256 |
| stream processor | 4096 | 2048 |
| Ray tracing unit | 32 | 16 |
| XMX engine | 512 | 256 |
| media engine | Four | 2 |
| GPU base clock | 1950MHz | 1600MHz |
| GPU max dynamic clock | 2050MHz | 1950MHz |
| memory | 16GB GDDR6 | 12GB GDDR6 (2x 6GB) |
| memory bus | 256 bits | 192 bits (2x 96 bits) |
| memory bandwidth | 576GB/s | 336GB/s |
| host interface | PCIe 4.0 x16 | PCIe 4.0 x 8 |
| TBP/TDP | 150W | 75W |
Both the card and GPU are powered by Intel’s Xe-HPG architecture. It is primarily designed with games in mind, but also artificial intelligence, Android and Windows cloud gaming, video transcoding (HEVC, AV1, AVC, and VP9 codecs), and virtual desktop infrastructure (VDI) applications. For now, Intel is positioning these boards for Android cloud gaming and media transcoding workloads. AI, VDI, and Windows cloud gaming will only be supported “when the product reaches full maturity.”
Intel’s Flex Series 170 (with one ACM-G10 GPU) can handle transcoding up to 8 simultaneous 4K video streams, rendering over 30 1080p streams, and 68 720p30 game streams. As a bonus, Intel’s high-end Arc Alchemist GPUs also support XMX instructions, which can speed up AI inference workloads.
In contrast, Intel’s dual-chip Flex Series 140 (with two ACM-G11 GPUs) delivers 8Kp60 real-time transformer with HDR in AV1 and HEVC/H.265 formats while meeting the industry’s 1 second latency requirement. I will give you the code. Also, a single card can handle 46 720p30 game streams (select titles only) and up to 216 game streams in multi-GPU configurations (I’m guessing 5 cards and 10 GPUs but this is a guess).
Intel’s Data Center GPU Flex family of server-grade graphics cards are fully supported by the company’s latest application programming interfaces and tools, including oneAPI, OpenVINO, oneVPL and VTune Profiler.
Servers with Intel’s Data GPU Flex family of cards are available from Dell, HPE, H3C, Inspur, Lenovo, and Supermicro.
