IBM and Together AI have announced a collaboration to deploy a dedicated Nvidia-powered inference cluster on IBM Cloud, a move that could reshape how enterprises access open-source AI services worldwide. The planned system is designed to improve performance, lower token costs, and expand global access to scalable AI infrastructure as demand for real-time applications grows.
Lumentum Holdings said it is pulling forward capacity investments across lasers, transceivers, and optical circuit switches (OCS) after fiscal fourth-quarter revenue surged 109% year-over-year to US$1.01 billion, with the company now guiding to hit its long-stated US$1.25 billion quarterly revenue target more than a quarter ahead of schedule.
Supermicro posted a sharp gross-margin recovery and issued fiscal 2027 revenue guidance well above Wall Street's expectations, addressing investor concerns that had built around whether the server maker's swelling AI order backlog could actually convert into profit.
Silicon Motion announced the pricing of a private placement of US$1 billion in aggregate principal amount of 0% convertible senior notes due 2031, increasing the fundraising size from the previously announced US$800 million.
July revenue trends across Taiwan's server supply chain remained broadly positive year over year, although sequential performance varied sharply by segment and supplier. ODM/EMS companies continued to generate the largest revenue base, while rail-kit maker King Slide, server manufacturer QCI and chassis supplier AIC recorded some of the strongest annual growth rates among the companies tracked.
MediaTek reported July 2026 revenue of NT$48.475 billion (US$1.5 billion), down 16.44% month-over-month but up 12.16% year-over-year, in line with its earlier conservative view that the third quarter would at best match the second quarter. The decline reflects weaker smartphone shipment momentum, even though the chipmaker still posted solid year-over-year growth.
Amid a swirl of rumors that Nvidia chip upgrades were being cut, scaled back, or delayed, the company has again pushed back on similar claims surrounding its latest Vera Rubin launch. Nvidia told Wccftech that its product strategy is not a downgrade but a flexible deployment model designed to meet different customer needs.
Global AI computing is rapidly moving beyond single-GPU systems toward AI Factory architectures, where agentic AI and inference workloads are pushing data centers to rely on networking, optics, packaging, cooling, power delivery and software co-design rather than raw accelerator counts. In a DIGITIMES interview, Nvidia Senior Vice President of Networking Gilad Shainer said the real unit of computing is now the entire data center, not an individual server.
