Qualcomm Solution for Inferencing | Solution

Efficient Inferencing with Qualcomm

The Qualcomm Cloud AI 100 achieves leading performance in throughput and latency while being as, or more, efficient as similar HHHL accelerators for smart edge processing. Additionally, the AI 100 shines with impressive efficiency in object detection and identification. This solution addresses the most important aspects of cloud AI inferencing—including low power consumption, scale, process node leadership, and signal processing expertise.

The GIGABYTE G292-Z43 was purpose-built for the AI 100 to support a high density of AI 100 cards. It also does so with optimal cooling that does not throttle performance. In the 2U chassis, it can support sixteen AI 100 cards and uses a dual AMD EPYC design.

Why Qualcomm Cloud AI 100 & GIGABYTE server?

High Performance

Up to 400TOPS per card and low-power consumption, 75W, from a single-slot GPU, the Qualcomm AI 100 provides a card for edge, telco, and datacenters.
Dense Accelerated Computing

Low-profile, HHHL, design allows for absurd GPU density using PCIe Gen4 lanes. The G292-Z43 supports up to 16 cards in x8 slots for high throughput.
Great Savings

Strong power efficiency translates into lowering TCO for edge and telco computing while the AI 100 is highly adaptable to environments.
Inference Efficiency

MLPerf v1.1 benchmark showcased leading performance by the Cloud AI 100 in inferences/second/watt. And also leadership in offline inferences/second.
Robust Software Tools

Major frameworks supported natively. Over 50 ML models for computer vision and natural language processing. Also python for development.
High Throughput

G292-Z43 can support sixteen Gen4 x8 lanes for AI 100 cards, and for faster networking there are an additional two low-profile Gen4 x16 lanes.

Cloud-Edge

Inference applications with leading hyperscale companies requiring heavy AI inferencing workloads for natural language processing, recommendations and prediction engines.
5G Infrastructure

Reshaping transportation and smart cities with massive MIMO (multiple-input, multiple-outfit) antennas.
5G Edge Box

Transforming the shopping experience, public safety, manufacturing (tracking defects), and agriculture.

Related Products

G292-Z43

HPC/AI Server - AMD EPYC™ 7002 - 2U DP 16 x PCIe Gen4 GPUs (Broadcom solution) | Application: AI , AI Training , AI Inference & HPC

Learn More

You May Be Curious

[Article] Qualcomm Cloud AI 100, AMD EPYC 7003 Series Processor, and Gigabyte server solutions breaks the Peta Operations Per Second barrier for AI Inferencing

[FAQ] Where can I buy a GIGABYTE server or motherboard?

Visit our Where to Buy page to learn about options based on regions across the world. Input your country, product, and type of seller. We have many ways to buy our products: GIGABYTE directly, a reseller, a system integrator, distributor, and more from our Where to Buy page. For further questions you can email us directly: marketing@gigacomputing.com

[FAQ] How long is GIGABYTE's server product warranty?

The warranty period and duration for GIGABYTE's server motherboards and barebone servers is dependent on many factors. Confirm with your seller. If it was GIGABYTE then contact us, if not then ask the seller.

[FAQ] If my GIGABYTE server products are out of warranty, are there any repair options available?

First, in order to confirm your product isn't still under warranty, you can submit an online RMA request and we can check for you. In the even that the warranty has expired, we do offer repairs at minimal costs. Visit our Service/Support site to learn more: https://www.gigabyte.com/Support

Telecommunication Visionary

In 1985 Qualcomm came to be, and made a big splash ten years later by developing Code Division Multiple Access (CDMA), which has been adopted worldwide for networking from 3G to present day 5G. Since then Qualcomm has been a pioneer in creating software and services for wireless technology and continues to improve the way we communicate.