Qualcomm, Meta collaborate to optimize Llama 3 to run on Snapdragon-powered devices

Qualcomm and Meta have joined forces to optimize Meta Llama 3 large language models (LLMs) to run on devices powered by upcoming Snapdragon platforms including smartphones, PCs, VR/AR headsets, vehicles, and more.

Developers will be able to access resources and tools in the Qualcomm AI Hub to run Llama 3 optimally on upcoming Snapdragon platforms, reducing time-to-market and unlocking on-device AI benefits.

Leveraging Qualcomm Technologies' powerful heterogeneous computing architecture, along with cutting-edge memory architecture, will allow customers, partners, and developers to maximize application performance, thermal efficiency, and battery life.

"We celebrate Meta's open approach with Meta Llama 3, and we share their commitment to empower developers and spur AI innovation. Our leadership in on-device AI, coupled with our vast reach across various devices at the edge, positions us to expand the benefits of the Llama ecosystem worldwide and enable customers, partners and developers to create a new generation of groundbreaking AI experiences," said Durga Malladi, senior vice president and general manager of technology, planning and edge solutions, Qualcomm Technologies, Inc.

The resources and tools to run Llama 3 will be accessible optimally on Snapdragon platforms via the Qualcomm AI Hub, which currently offers approximately 100 optimized AI models. The Qualcomm AI Hub reduces time-to-market for developers and unlocks the benefits of on‑device AI for their apps.

The partnership between Meta and Qualcomm is a significant milestone in their ongoing collaboration aimed at empowering developers and users with powerful AI capabilities.