Qualcomm offers test drive of its rack-scale Nvidia AI challenger
What happened
Qualcomm has opened cloud-based trials of its Dragonfly AI200 rack-scale inference system to selected customers through Cirrascale Cloud Services, allowing customers to test the system on Cirrascale's AI Innovation Cloud, with what Qualcomm says is a route into production deployments. Cirrascale also demonstrated the system at SC26 in Chicago on November 17-18. The AI200 is part of Qualcomm's Dragonfly datacenter lineup, which also includes the AI250 and AI300 accelerators and the C1000 server CPU. Each AI200 accelerator card carries 768 GB of LPDDR memory, with 56 cards providing about 43 TB in a liquid-cooled rack. Qualcomm says the system supports models ranging from 7 billion to 10 trillion parameters, including long-context, retrieval-augmented generation (RAG), and agentic workloads. The source does not state the cause.
Impact and signal
Affected components and suppliers
Source and time
Source: The Register · 11/19/2026 · View original source
Real-time pricing, supply signals and AI analysis for semiconductor traders