Nokia strengthens edge AI capabilities through collaboration with Blaize in Asia Pacific

Nokia has entered into a strategic collaboration with US-based AI computing company Blaize to advance hybrid inference solutions across Asia Pacific, as enterprises increasingly look to deploy artificial intelligence reliably in real-world environments. The partnership is formalised through a non-binding Memorandum of Understanding between Blaize and Nokia Solutions and Networks Singapore.

The agreement focuses on developing Practical AI and Physical AI systems that can operate efficiently at the network edge while remaining integrated with cloud and data centre infrastructure. By combining Nokia’s strengths in networking, automation, and cloud technologies with Blaize’s energy-efficient AI inference platform, the companies aim to address growing demand for low-latency, resilient AI deployments across industries where performance and power efficiency are critical.

Targeted at Asia Pacific markets, the collaboration reflects a broader industry shift away from model training alone towards inference at scale. Enterprises are increasingly seeking architectures that balance centralised cloud resources with intelligence deployed closer to where data is generated and decisions are executed, particularly in telecoms, industrial automation, and smart infrastructure environments.

Focus on hybrid inference and real-world deployment

Under the framework outlined in the MOU, Nokia and Blaize plan to jointly explore and validate edge and hybrid AI inference use cases. This includes integrating Blaize’s hybrid AI platform with Nokia’s IP networking, data centre networking, and automation systems to support AI workloads across distributed environments.

The collaboration also covers the development of reference architectures and solution blueprints, positioning Blaize as a complementary inference layer alongside Nokia’s existing AI networking infrastructure. These designs are intended to help customers deploy production-ready AI systems that can scale reliably while maintaining operational efficiency.

In addition, the companies plan to work together on validating AI inference deployments across sectors such as telecommunications, industrial operations, and smart infrastructure. These environments often demand predictable performance, low power consumption, and the ability to operate under constrained or variable network conditions, making hybrid architectures particularly relevant.

Industry context and executive perspectives

The collaboration is underpinned by a shared view that the next phase of AI adoption will be driven by inference rather than training. As AI systems mature, organisations are placing greater emphasis on where inference runs and how it integrates with existing infrastructure, especially as cloud costs, latency requirements, and sustainability considerations come under closer scrutiny.

Dinakar Munagala, Co-Founder and Chief Executive Officer of Blaize, said the partnership represents a step forward in deploying AI at scale in real-world settings. “By combining Nokia’s leadership in connectivity and automation with the Blaize AI inference platform, we are enabling Real World AI that runs efficiently at the edge while integrating seamlessly with cloud and GPU infrastructure. This Hybrid AI approach allows organisations to deploy inference where it matters most and turn intelligence into real operational outcomes.”

From Nokia’s perspective, the collaboration aligns with its focus on extending AI capabilities across networks, edge systems, and cloud environments. Sang Xulei, Senior Vice President and Head of Network Infrastructure, Asia Pacific at Nokia, highlighted the growing demand for AI systems that can operate outside controlled data centre settings. “As demand grows for Practical AI and Physical AI systems that operate in real world environments, Blaize’s energy efficient AI inference platform gives us the flexibility to extend Hybrid AI architectures across networks, edge systems, and cloud infrastructure.”

Scope and outlook for the Asia Pacific region

The MOU sets out a cooperative framework rather than binding commercial commitments, with specific projects to be pursued through future definitive agreements. The collaboration will focus on secure, scalable, and energy-efficient AI inference deployments that can integrate into existing enterprise, network, and industrial environments across Asia Pacific.

Blaize’s platform has been selected for its ability to support AI inference across edge, cloud, and data centre environments, enabling applications such as computer vision, multimodal AI, and sensor-driven systems. These capabilities are increasingly relevant to smart cities, telecommunications, logistics, and industrial automation use cases across the region.

For Nokia, the partnership reinforces its positioning as a connectivity provider for the AI era, where networks are expected to support not just data transport but also intelligent processing closer to the point of action. As enterprises across Asia Pacific continue to move beyond pilot projects, collaborations focused on operational AI deployment are likely to play a growing role in shaping how AI is delivered at scale.