Oppo, a major player in smart devices, has introduced an innovative AI (artificial intelligence) model called the Mixture of Experts (MoE) architecture, directly integrated into its devices. This move positions Oppo as the first company to bring MoE’s efficiency-focused AI framework directly onto mobile hardware, setting the stage for enhanced AI performance in smartphones and related devices.
AI models have traditionally demanded high computational power, which can strain device resources, especially in compact hardware. In partnership with major chipset makers, Oppo’s MoE architecture addresses this challenge by selectively activating specialized sub-models, or “experts,” for specific tasks, which conserves processing power and minimizes data transfer.
Lab tests show that using the MoE model improves AI response times by around 40%, while also reducing energy use. For users, this translates into quicker AI responses, longer battery life, and better privacy, as more data processing happens locally on the device, limiting the need for cloud-based processing.
Oppo’s adoption of the MoE architecture shows a shift in mobile AI capabilities, opening up the possibility for even budget-friendly devices to handle complex AI tasks effectively. This approach aligns with Oppo’s broader goal of making AI-powered technology available to more users across different price ranges.
“With over 5,860 AI-related patents, Oppo is heavily invested in the future of AI,” the company said in a statement. “The establishment of its AI Center this year underscores our commitment to research, pushing forward advancements that make AI more efficient and accessible.”
Image from Oppo Store Facebook page.