Qualcomm unveils Snapdragon Reality Elite for smart glasses

๐กNew silicon for XR devices is the backbone of the next wave of spatial AI and AR applications.
โก 30-Second TL;DR
What Changed
New Snapdragon Reality Elite chip targets XR and smart glasses
Why It Matters
Enhanced mobile silicon for XR will accelerate the development of lightweight, AI-capable smart glasses, bridging the gap between mobile and spatial computing.
What To Do Next
If building spatial applications, benchmark your rendering pipelines against the new Snapdragon Reality Elite specifications.
๐ง Deep Insight
Web-grounded analysis with 12 cited sources.
๐ Enhanced Key Takeaways
- โขThe Snapdragon Reality Elite platform delivers a 160% improvement in NPU performance and a 30% increase in CPU performance compared to the previous Snapdragon XR2 platform.
- โขIt features an on-device AI performance of up to 48 TOPS, enabling direct support for large language models (LLMs) and vision AI models on XR devices, facilitating features like AI assistants and photorealistic avatars.
- โขThe chip supports displays with up to 4.4K resolution per eye at 90 frames per second, aiming for sharper visuals and smoother mixed reality experiences.
- โขAn integrated Engine for Visual Analytics (EVA) block provides hardware acceleration for demanding computer vision workloads, enhancing capabilities such as depth estimation, head, and hand tracking.
- โขBeyond smart glasses, the Reality Elite is designed to power a wide range of XR form factors, including all-in-one mixed reality devices and tethered headsets.
๐ ๏ธ Technical Deep Dive
- GPU Performance: Up to 60% higher Adreno GPU performance compared to Snapdragon XR2+ Gen 2.
- CPU Performance: Up to 30% faster Kryo CPU performance, featuring a 4 + 2 performance core configuration.
- NPU Performance: Up to 160% higher Hexagon NPU performance, delivering 48 TOPS for AI processing.
- Display Support: Capable of supporting up to 4.4K resolution per eye at 90 FPS.
- Power Efficiency & Thermal Management: Offers up to 20% longer battery life and runs up to 12 degrees Celsius cooler under load compared to XR2+ Gen 2.
- AI Capabilities: Supports on-device execution of 3 billion parameter LLMs at 45 tokens per second and 512x512 large vision models with approximately 1.7 seconds of latency.
- Computer Vision Acceleration: Includes an Engine for Visual Analytics (EVA) block for hardware-accelerated computer vision tasks, improving depth estimation and tracking.
- Memory & Storage: Supports faster UFS 4.0 storage and 4.2 GHz RAM (up from 3.2 GHz), with a 30% increase in DDR bandwidth.
- Connectivity: Features built-in support for up to 2x USB 3.1 ports and Bluetooth 6.0.
๐ฎ Future ImplicationsAI analysis grounded in cited sources
โณ Timeline
๐ Sources (12)
Factual claims are grounded in the sources below. Forward-looking analysis is AI-generated interpretation.
Weekly AI Recap
Read this week's curated digest of top AI events โ
๐Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: The Verge โ

