ASUS Takes the Lead in Hybrid Agentic AI Infrastructure
As enterprises accelerate the adoption of large language models (LLMs) and AI agent-based applications, inference-related token costs continue to rise. Pure cloud-based approaches not only introduce high and unpredictable costs, but also create operational challenges, making cost efficiency a key barrier to large-scale AI deployment.
To address these challenges, ASUS introduces a hybrid AI architecture that dynamically distributes workloads between local devices and the cloud. By handling a portion of AI tasks locally and reserving the cloud for more complex workloads, the architecture maintains performance while significantly reducing overall inference costs and improving deployment flexibility across different device form factors.
“Balancing performance and cost has become a critical challenge as enterprises scale their AI adoption,” said Bryan Chang, General Manager of Commercial PC BU, ASUS. “With hybrid AI, we aim to shift more AI processing to the device, reducing reliance on cloud resources and improving both efficiency and practicality. Expanding this capability across laptops, desktops, and mini PCs reflects our commitment to evolving AI PCs from standalone devices into scalable, real-world solutions.”
At the technology level, the architecture integrates Phison’s aiDAPTIV™ memory extension technology, enabling devices with limited hardware resources to support mid- to large-scale language models locally. This removes traditional memory constraints and allows AI workloads previously requiring high-end infrastructure to run on commercial PC platforms. In addition, a gateway-based routing mechanism intelligently assigns tasks based on complexity, prioritizing local processing where possible to further optimize efficiency.
According to results from PinchBench, a benchmarking system for evaluating LLM models as OpenClaw coding agents, the hybrid inference approach can reduce inference costs for mid- to large-scale models (such as 26B and 35B) by up to 70% while maintaining performance, offering a more sustainable and cost-efficient model for enterprise AI deployment.
“aiDAPTIV enables local execution of larger AI models by overcoming traditional memory limitations, while hybrid inference significantly reduces overall compute costs,” said K.S. Pua, CEO of Phison. “Our collaboration with ASUS demonstrates how this technology can be effectively deployed across commercial platforms, delivering a scalable and cost-efficient solution for enterprise AI adoption.”
With a diverse portfolio spanning laptops, desktops, and mini PCs, ASUS is extending hybrid AI capabilities across a wide range of enterprise use cases. These include multilingual translation, business email drafting, meeting note summarization, contract and long-document summarization, internal knowledge base Q&A, automated customer service and FAQ responses, as well as CRM record management and sales support. By handling these tasks locally, enterprises can reduce dependence on cloud token consumption while improving data privacy and response efficiency, creating a more cost-effective approach to AI deployment. Combined with scenarios such as on-device AI assistants, log analysis, and troubleshooting support, the hybrid AI architecture enables organizations to deploy AI more flexibly according to their operational needs and accelerate the transition from pilot projects to large-scale implementation.
Looking ahead, ASUS will continue to advance its AI PC and commercial computing platforms by integrating hardware, software, and ecosystem partnerships. The company aims to deliver AI infrastructure that balances performance, cost, and flexibility, enabling enterprises to scale AI adoption and drive long-term digital transformation.
Did this article help you? If so, please tell me in a comment what do you think about it.
Don’t miss any of our future video tutorials, follow us on Youtube. Like us on Facebook. Join our Best Deals Telegram Channel. Join our Android TV Box Firmware Updates Telegram Channel. Subscribe now to our newsletter. If you need Tech Reviewer or Youtube Influencer read this. Donate now here to support CGR Team!
