---
title: "AI200"
url: "https://www.qualcomm.com/data-center/products/qualcomm-dragonfly-ai200"
description: "Qualcomm Dragonfly AI200: a rack-scale AI inference accelerator with 43TB rack memory, 768GB per card, optimized for generative and agentic AI at scale."
---

# AI200

Qualcomm Dragonfly™

Active

![chip image](https://s7d1.scene7.com/is/image/dmqualcommprod/Dragonfly_AI200_Raster_RGB_Color?fmt=webp-alpha&qlt=85)

[Contact Sales](https://www.qualcomm.com/support/contact/forms/data-center)

On this page

[Overview](https://www.qualcomm.com/#overview)
[Benefits](https://www.qualcomm.com/#benefits)
[Features / Specs](https://www.qualcomm.com/#features)

Purpose-built rack-scale AI inference with the industry's highest memory capacity per accelerator¹ — engineered for generative and agentic AI at scale.

Qualcomm Dragonfly™ AI200 is a purpose-built, rack-scale AI inference accelerator engineered for the generative and agentic AI era. Each AI200 card delivers an industry-leading 768 GB of LPDDR memory¹, while a single 140 kW liquid-cooled ORv3-compliant rack provides industry-leading 43 TB of memory capacity¹  — enabling efficient inference up to 10T parameter models with compelling total cost of ownership (TCO).

Product license agreement

## Benefits

- **Best-in-class Memory Capacity¹**
With 768 GB per card and 43 TB per rack, AI200 serves models from 7B up to 10 trillion parameters, supporting long-context, RAG, and agentic workloads without compromise.

- **Rack-Scale Deployment**
A fully integrated OCP ORv3-compliant, direct-liquid-cooled rack with cableless backplane simplifies deployment, accelerates time-to-production, and improves reliability across enterprise and hyperscale data centers.

- **Leading Total Cost of Ownership**
Purpose-built for inference economics, AI200 delivers compelling performance-per-dollar, driving down cost-per-token for generative and agentic AI workloads.

- **Complete Inference Software Stack**
The Qualcomm® AI Inference Suite provides end-to-end tooling for model deployment as bare-metal, cloud VM, or inference-as-a-service — reducing complexity and accelerating time-to-value.

- **Direct Liquid-Cooled**
A 140 kW direct-liquid-cooled OCP ORv3-compliant rack maximizes compute density and energy efficiency, lowering power usage effectiveness (PUE) and enabling sustainable AI infrastructure at scale.

¹Compared to commercially-available competing published product specifications.

## Features

- Rack-scale solution purpose-built for AI inference economics 
- 768 GB LPDDR memory per card — highest per-accelerator capacity in the industry¹
- Industry-leading 43 TB memory capacity per rack¹
- Supports SOTA models up to 10T parameters with context lengths up to 128K 
- PCIe Gen6 scale-up; Ethernet-with-RoCE scale-out 
- Qualcomm® AI Inference Suite for bare-metal, VM, or IaaS deployment 
- Designed to accelerate various model categories - reasoning, vision, agentic, LLM, text-to-image, and video processing 
- Air and direct liquid cooling
- OCP ORv3-compliant rack with cableless backplane

### Specifications

| Category | Attribute | Value |
| --- | --- | --- |
| Rack | Form Factor | Single Wide, Open Rack v3 (ORv3) compliant |
| Rack | Number of Cards | 56 |
| Memory | Bandwidth | 0.414 PB/s[1](https://www.qualcomm.com/#footnote1) |
| Memory | Capacity | 43 TB |
| Scale-Up | Interface Type | PCIe 6.0 |
| Scale-Out | Interface Type | Ethernet w/RoCE |
| Thermal Management | Cooling | Direct Liquid Cooling (DLC), Air Cooling |
| Thermal Management | Thermal Design Power | 140 kW |

1. With LPDDR5x

![](https://s7d1.scene7.com/is/image/dmqualcommprod/2000_?fmt=webp-alpha&qlt=85)

**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. 

[Learn more](https://www.qualcomm.com/data-center/solutions/ai-infrastructure-management-suite)