2
Image may differ from actual product
Qualcomm Cloud AI 100 Accelerator for HPE
Does your organization require optimized infrastructure for deploying AI/ML trained models at the edge? The Qualcomm Cloud AI 100 Accelerator for HPE dramatically improves server performance on AI/ML inference workloads for faster insights and higher accuracy data transformations on computer vision, natural language, and other state-of-the-industry AI/ML use cases. Based on the Qualcomm AI Core, this server accelerator is tailored for inference, supporting model customization and quantization to lower precision datatypes for optimized performance and industry-leading power efficiency. With low power consumption, the Qualcomm Cloud AI 100 Accelerator for HPE can increase performance while lowering operating cost.
SKU # R9Q09A
Get Started
- QuickSpecs
-
Existing selections will be lost. Click OK to proceed further.
More Information
What's New
- Compact, single-slot, low-profile, and low-power PCIe accelerators is optimized for AI interference infrastructure at the edge.
- Based on Qualcomm AI Core, Inference Optimized Architecture leverages over a decade of power-efficient deep learning technology for inference computing.
- Achieves impressive peak TOPS (trillion operations per second) performance of 350 Int8, 175 FP16.
- High performance can dramatically lower infrastructure acquisition costs.
Key Features
High Performance
Purpose-designed for AI/ML workloads, the Qualcomm Cloud AI 100 Accelerator for HPE is a market leader in performance and power efficiency based on industry-standard benchmarks including MLPerf.
Up to eight of these low-profile PCIe accelerators can be configured within an HPE Edgeline EL8000.
Each standard card delivers 350 TOPS while consuming only 75 watts of power.
Using ResNet-50 benchmark, performance translates to inferencing up to over 100,000 images per second in a single HPE Edgeline EL8000 system.
Compact and Energy Efficient
Each Qualcomm Cloud AI 100 Accelerator for HPE is rated at 75 Watts TDP.
Qualcomm Cloud AI 100 Accelerator for HPE design is a PCIe Gen4 (x8) connector in a low-profile form factor (both half-height and half-length) for compact use at the edge.
Easy to Use
The Qualcomm Cloud AI 100 Accelerator for HPE leverages Qualcomm's 6th generation of AI/ML technologies and includes a comprehensive SDK and extensive suite of developer tools (compiler, performance optimization, and quantization tools) and code samples.
Features simplified support for popular AI/ML frameworks.
* Prices may vary based on local reseller.
Find what you are looking for?
Need help locating the right product for your business?
Our product experts would love to chat with you to find products and services that enable opportunities and resolve your business problems.