Qualcomm to buy Modular in push for edge AI software

Qualcomm has agreed to acquire AI software company Modular, adding its platform to Qualcomm's push across edge and data centre AI.

The acquisition is intended to strengthen Qualcomm's software base for generative and agentic AI in both environments. Modular's platform is expected to help Qualcomm support developers, hardware makers, cloud providers and model creators working across different computing systems.

Modular has built an AI software stack designed to run models across several hardware architectures, including CPU, GPU, NPU and custom ASIC systems, without requiring developers to rewrite code for each type of accelerator. Qualcomm said that approach would support a broader software layer that works across devices, edge systems and data centres.

The transaction also extends Qualcomm's data centre ambitions. The addition of Modular is expected to deepen the software side of its data centre strategy, including inference, orchestration and deployment in distributed AI systems.

That matters because the market for AI infrastructure is increasingly shaped by efficiency and deployment costs rather than raw model size alone. Qualcomm argued that developers need software that can translate chip-level performance into practical services across a mix of hardware and operating environments.

Modular was founded by engineers involved in building parts of today's AI infrastructure. Its platform has been positioned around hardware portability and a vendor-neutral approach, allowing developers to build once and deploy in different environments.

Software focus

For Qualcomm, the deal signals a stronger emphasis on software as it seeks to expand beyond its established semiconductor business. The company has long been associated with mobile and connected computing chips, but has been widening its AI strategy across on-device systems, edge computing and, more recently, data centre workloads.

By bringing in Modular's engineering team and platform, Qualcomm is aiming to make its hardware more accessible to developers earlier in deployment cycles. It also wants to support systems that combine chips and accelerators from more than one vendor.

Cristiano Amon, President and Chief Executive Officer of Qualcomm, framed the transaction as part of a broader shift in AI computing architecture.

"This acquisition marks a pivotal moment not just for Qualcomm, but for the AI industry," said Cristiano Amon, President and Chief Executive Officer of Qualcomm. "As agentic AI scales across data centers and edge environments, the industry is moving toward disaggregated, multi-vendor architectures that demand a more open and modern software foundation. We believe the future belongs to developer-friendly, horizontal platforms that can run across diverse compute environments and give customers real choice in how and where they deploy AI. With Modular, we're accelerating that shift, combining our scale and energy-efficient data center technologies with an open ecosystem approach to help drive the next chapter of AI."

Open ecosystem

Qualcomm said the combination would help it build a more open developer ecosystem across varied computing environments. It described Modular's community as vendor-neutral and focused on improving portability and efficiency in AI infrastructure.

That reflects a broader debate in the AI market over whether customers should tie software stacks to a single hardware supplier or use tools that can move workloads between different systems. Qualcomm is positioning itself on the side of portability as competition grows in both edge AI and data centre inference.

Chris Lattner, Co-founder and Chief Executive Officer of Modular, said the company saw the deal as a way to extend that approach.

"Modular was founded on the belief that AI needs a more open and efficient software foundation that can span diverse hardware and deployment environments," said Chris Lattner, Co-founder and Chief Executive Officer of Modular. "Joining Qualcomm gives us the scale and platform reach to accelerate that mission. Together, we can make AI development more accessible and performant for developers, strengthen portability across hardware, and help grow an open ecosystem that broadens participation and speeds innovation. We are excited to continue advancing our software platform as part of Qualcomm's broader strategy from edge to cloud."

No financial terms were disclosed. The transaction is expected to close in the second half of 2026, subject to customary closing conditions and regulatory approvals.

The deal gives Qualcomm a software asset spanning several layers of AI deployment, from model execution to workload management across mixed hardware. It also strengthens the company's position with developers and enterprise customers deploying AI systems across both devices and larger-scale infrastructure.

Modular's technology is aimed at reducing the need for separate optimisation work across different chip types, a problem that has become more pressing as AI workloads spread beyond centralised cloud systems. Qualcomm said that portability could lower total cost of ownership for customers deploying AI across a mix of environments.

The acquisition comes as chip companies seek tighter control over the software that determines how efficiently AI models run on their hardware. For Qualcomm, buying Modular adds a layer intended to link its semiconductor strategy more closely with the tools developers use to build and deploy AI services across edge and cloud systems.