Qualcomm Dragonfly AI300
Active
Qualcomm Dragonfly™
AI300
Contact Sales
chip image

Third-generation, rack-level AI inference platform integrating our next-gen near-memory computing with increased bandwidth scale-up and scale-out, designed for disaggregated inference deployments

Qualcomm Dragonfly™ AI300 expands the Qualcomm Dragonfly AI inference offering by integrating Qualcomm® Hign Bandwidth Compute (HBC) Gen 2, enabling full all-to-all rack level scale-up, and high bandwidth scale-out to support full disaggregated inferencing deployments. The solution will expand from a rack to a pod offering supporting the entire inference pipeline and enables fast deployment.

Built for hyperscale AI, it is designed for industry-leading1 memory capacity and effective bandwidth enabling high-throughput, low-latency performance for large language & multimodal model(LMM, LMM) inference and agentic AI workloads.

Product license agreement

Benefits

default icon alt text
Industry-leading Perf/TCO for LLM Decode¹
AI300 with HBC Gen 2 is designed for industry-leading memory capacity and effective memory bandwidth, offering the right balance to efficiently service large context attention as well as large model feed forward networks (FFN).
default icon alt text
Standardized Pod Configurations
AI300 fully validated pod configurations enable the capability to run end-end inference pipelines in a self contained unit of deployment
default icon alt text
AI Inference Software
Expands our AI inference capabilities to deliver seamless disaggregated inference solutions across Qualcomm Technologies and other vendor hardware.
default icon alt text
Power-efficient Inference
Qualcomm HBC with near-memory computing architecture reduces data movement — the dominant source of energy consumption in inference — increasing tokens-per-watt.
default icon alt text
Holistic co-designed System Solution
Custom CPU, AI accelerator, connectivity, software, and algorithms are developed simultaneously from the ground up for AI inference.

1Compared to competing published product specifications normalized at card- and rack-level.

Features

  • AI300 accelerator cards with fully integrated compute, memory and networking on a single direct liquid cooled (DLC) rack
  • Integrated Qualcomm HBC Gen 2 is engineered for industry-leading effective memory bandwidth 1
  • All-to-all rack scale-up with a single hop to enable high bi-directional bandwidth per AI300 accelerator
  • Qualcomm Dragonfly™ C1000 CPU delivers high per core performance for accelerating agentic AI use-cases
  • Expanded scale-out per AI300 card to enable low latency full prefill-decode disaggregation at scale
  • Fully validated pod configurations to enable complete end-end inference solution for customers
  • Air and direct liquid cooling

QUALCOMM DRAGONFLY AI INFRASTRUCTURE MANAGEMENT SUITE

Orchestrate and
scale efficiency

The suite provides provisioning, monitoring, orchestration and fault handling across rack‑scale deployments. Together, hardware, connectivity and software form the foundation of a cohesive data center platform approach — one designed to scale with customers as AI workloads evolve. 

Qualcomm relentlessly innovates to deliver intelligent computing everywhere, helping the world tackle some of its most important challenges. Our leading-edge AI, high performance, low-power computing, and unrivaled connectivity deliver proven solutions that transform major industries. At Qualcomm, we are engineering human progress.

Stay connected

Get the latest Qualcomm and industry information delivered to your inbox.

Subscribe
Manage your subscription
Also of Interest

© Qualcomm Technologies, Inc. and/or its affiliated companies.

Snapdragon and Qualcomm branded products are products of Qualcomm Technologies, Inc. and/or its subsidiaries. Qualcomm patented technologies are licensed by Qualcomm Incorporated.

Note: Certain services and materials may require you to accept additional terms and conditions before accessing or using those items.

References to "Qualcomm" may mean Qualcomm Incorporated, or subsidiaries or business units within the Qualcomm corporate structure, as applicable.

Qualcomm Incorporated includes our licensing business, QTL, and the vast majority of our patent portfolio. Qualcomm Technologies, Inc., a subsidiary of Qualcomm Incorporated, operates, along with its subsidiaries, substantially all of our engineering, research and development functions, and substantially all of our products and services businesses, including our QCT semiconductor business.

Materials that are as of a specific date, including but not limited to press releases, presentations, blog posts and webcasts, may have been superseded by subsequent events or disclosures.

Nothing in these materials is an offer to sell or license any of the services or materials referenced herein.