IT Brief New Zealand - Technology news for CIOs & IT decision-makers
New Zealand
Qualcomm & Cirrascale open Dragonfly AI200 for tests

Qualcomm & Cirrascale open Dragonfly AI200 for tests

Wed, 18th Nov 2026 (Today)
Joseph Gabriel Lagonsin
JOSEPH GABRIEL LAGONSIN News Editor

Qualcomm and Cirrascale have made the Qualcomm Dragonfly AI200 available for evaluation by select customers through Cirrascale's AI Innovation Cloud, giving them access to the companies' latest AI inference infrastructure collaboration.

Users can test the system with their own models and workloads in Cirrascale's cloud environment. The setup is intended to support platform validation, workload characterisation, benchmarking and proof-of-concept work, with a route to production inference deployments.

The announcement focuses on AI inference, an area drawing more attention as companies shift from training models to running them at scale in live services. Qualcomm and Cirrascale are positioning the Dragonfly AI200 as a rack-scale option for large language models, multimodal systems and other AI workloads where power use and operating costs are under scrutiny.

The hardware supports 768GB of LPDDR memory per card. Qualcomm says the platform is architected for inference at trillion-parameter scale and combines that memory capacity with rack-level design features including direct liquid cooling, PCIe scale-up, Ethernet scale-out and confidential computing.

Cirrascale is providing the cloud environment for the evaluations. Its service includes managed AI infrastructure and a flat-rate commercial model without ingress or egress fees, giving customers a dedicated setting to assess systems designed specifically for inference rather than general-purpose infrastructure.

Dave Driggers, Chief Executive Officer of Cirrascale Cloud Services, outlined the company's view of where demand is moving in AI infrastructure.

"The economics of AI are increasingly becoming the economics of inference. Our customers need efficient, scalable infrastructure that can support increasingly sophisticated models without allowing power and operating costs to become the limiting factor. By working with Qualcomm Technologies, we can give customers evaluation access to a compelling new architecture purpose-built for the next generation of AI inference," said Driggers.

Inference focus

The partnership reflects a broader shift in the AI market. While much of the sector's spending has focused on model development and training, suppliers are now competing to address the cost and energy demands of deploying those models at data-centre scale.

Qualcomm, better known for mobile and edge computing, has been expanding its push into data-centre AI inference. The Dragonfly AI200 is part of that effort, which the company describes as built around the memory, performance and cost profile needed to serve AI models rather than train them.

Tony Pialis, Executive Vice President and General Manager, Data Centre, Qualcomm Technologies, set out the case for a dedicated inference architecture.

"AI inference is becoming one of the largest and most important computing workloads in the data center, and the industry needs a more efficient approach. Qualcomm AI200 brings our leadership in high-performance, low-power AI computing to rack-scale infrastructure, with an architecture designed around the performance, memory and cost requirements of inference. Our work with Cirrascale gives customers a direct way to evaluate this technology on their own workloads and experience the resulting data center economics," said Pialis.

Cloud route

For Cirrascale, the arrangement adds a new system to its AI Innovation Cloud at a time when cloud providers and infrastructure specialists are trying to attract customers seeking alternatives to mainstream compute platforms. By offering evaluation access in a managed environment, Cirrascale is aiming to lower the barrier for enterprises and cloud users that want to compare inference systems on real workloads before committing to deployment.

Select customers can move from evaluation to production inference deployments through the same route. That matters in a market where many AI buyers want faster ways to test performance, energy use and operational cost before scaling workloads.

Qualcomm also says the system is supported by its AI Management Suite and ecosystem support for leading AI frameworks. That software layer is intended to help customers assess how the platform handles production-style inference tasks alongside the underlying hardware design.

The collaboration also signals longer-term intent from both companies in AI infrastructure. Qualcomm says the work with Cirrascale could serve as a base for further work across its multi-generation data-centre AI inference roadmap.

Dragonfly AI200 is available now for evaluation by select customers through Cirrascale.