Broadcom Inc. reports developments across semiconductor and infrastructure software businesses serving enterprise, cloud, broadband and AI infrastructure markets. Company news regularly covers custom silicon and networking technology for AI compute, Ethernet-based data-center connectivity, Wi-Fi and PON broadband chips, and infrastructure software products built around VMware Cloud Foundation, Tanzu, AppNeta and Arcot.
Recurring updates also include quarterly financial results and business outlooks, strategic customer and cloud collaborations, product releases for private cloud, observability and payment authentication, and governance changes within the Delaware corporation.
Broadcom (NASDAQ: AVGO) has announced details for VMware Explore 2025, a major technology conference scheduled for August 25-28 at The Venetian in Las Vegas. The event will feature 400+ technical sessions across four tracks focusing on cloud infrastructure, private AI, modern applications, and networking.
Broadcom CEO Hock Tan will deliver a keynote on August 26 titled "Shaping the Future of Private Cloud and AI Innovation." The conference includes a redesigned Expo merged with The Hub, offering hands-on labs, certification training, and networking opportunities. Following the Las Vegas event, VMware Explore will tour globally with events in Mumbai, London, Paris, Sydney, Tokyo, and Frankfurt from September to November 2025.
[]Broadcom (NASDAQ: AVGO) has announced the shipping of Jericho4, a groundbreaking ethernet fabric router designed for distributed AI infrastructure. This new platform can interconnect over 1 million XPUs across multiple data centers, featuring 3.2 Tbps HyperPort technology and the ability to scale to 36,000 HyperPorts.
The Jericho4, manufactured on a 3nm process, offers significant improvements including deep buffering, line-rate MACsec encryption, and lossless RoCE transport over 100km+ distances. The platform's HyperPort technology boosts utilization by up to 70% and eliminates load balancing inefficiencies, while supporting full MACsec encryption on every port.
Broadcom (NASDAQ: AVGO), a global technology leader in semiconductor and infrastructure software solutions, has scheduled its third quarter fiscal year 2025 financial results announcement for Thursday, September 4, 2025, after market close.
The company will host a conference call at 2:00 PM Pacific Time (5:00 PM Eastern Time) on the same day. Investors can access the live webcast through Broadcom's investor relations website, where an audio replay will be available for one year.
Broadcom (NASDAQ:AVGO) has announced the shipment of Tomahawk Ultra, a groundbreaking Ethernet switch designed specifically for high-performance computing (HPC) and AI workloads. The switch delivers ultra-low latency of 250ns at full 51.2 Tbps throughput and supports up to 77 billion packets per second.
Key innovations include reduced Ethernet header overhead from 46 to 10 bytes, lossless fabric technology through Link Layer Retry and Credit-Based Flow Control, and In-Network Collectives for accelerated AI operations. The company also introduced SUE-Lite, an optimized version of Scale-Up Ethernet for power-sensitive accelerator applications.
Tomahawk Ultra is 100% pin-compatible with Tomahawk 5 and is currently shipping for deployment in rack-scale AI training clusters and supercomputing environments.
Broadcom (NASDAQ: AVGO) has unveiled its third-generation co-packaged optics (CPO) technology featuring 200G per lane capability, marking a significant advancement in AI networking solutions. This announcement builds on Broadcom's successful CPO journey, which began in 2021 with the first-generation Tomahawk 4-Humboldt chipset and progressed to the second-generation Tomahawk 5-Bailly volume-production solution.
The company has demonstrated ecosystem readiness through key partnerships with Corning, Delta Electronics, Foxconn Interconnect Technology, Micas Networks, and Twinstar Technologies. These collaborations have resulted in significant improvements in OSAT processes, thermal designs, and overall manufacturing yield. The new 200G/lane CPO technology is specifically designed for next-generation high-radix scale-up and scale-out networks, crucial for enabling scale-up domains exceeding 512 nodes.