📋About Broadcom
Updated September 16, 2026Broadcom Inc. is a global technology leader that designs, develops, and supplies a broad range of semiconductor, enterprise software, and security solutions. Following its landmark $69 billion acquisition of VMware in 2023, Broadcom became one of the largest infrastructure technology companies in the world, with a combined portfolio spanning networking, storage, virtualization, and cybersecurity.
In the AI space, Broadcom plays a critical infrastructure role. The company's custom silicon division co-developed Jalapeño, OpenAI's first custom AI chip — a reticle-sized inference ASIC unveiled in June 2026 and detailed at the Hot Chips conference on August 25, 2026. It was built from a blank sheet for language-model inference and uses HBM4 memory. OpenAI invited SemiAnalysis into its lab to run the InferenceX benchmark suite, the first third-party measurement of the part: it recorded more than 700 tokens per second per user on DeepSeek R1 at a concurrency of one, without speculative decoding, and better throughput per watt than the Nvidia Blackwell system it was compared against. Two caveats travel with those figures, both stated by SemiAnalysis — OpenAI supplied the numbers, and the like-for-like comparison is Nvidia's shipping Rubin generation rather than Blackwell, since Rubin also uses HBM4. Deployment begins at the end of 2026 in very small volumes, with meaningful volume expected in 2027. Broadcom also designs custom AI accelerators (XPUs) for other hyperscale customers including Google (TPU) and Meta, positioning the company as the leading merchant designer of custom AI chips.
Beyond silicon, Broadcom's networking portfolio — including Memory Interconnect, Tomahawk and Jericho switching ASICs, and 800G/1.6T Ethernet solutions — forms the backbone of AI data center interconnects. VMware's Private AI Foundation and Tanzu platforms further extend Broadcom's reach into enterprise AI infrastructure, enabling organizations to run AI workloads on private cloud environments with GPU virtualization and Kubernetes orchestration.
🛠️Products & Tools (2)
Industry-leading 51.2 Tbps Ethernet switch ASIC powering AI data center fabrics. Supports 800GbE ports for GPU cluster interconnects and large-scale AI training networks.
Enterprise platform for deploying generative AI on private infrastructure. Integrates GPU virtualization, Kubernetes orchestration via Tanzu, and curated open-source LLMs from NVIDIA and partners.
Keep track of the companies you’re watching
- Save the companies you want to follow
- Get ⚡ alerts when your saved companies change
- Every product they ship, cross-linked to 900+ AI tool profiles
- Today’s top AI Stories — the day’s most important AI news, free
Swipe for Recommended for you and My AI Tools
Your AI Hub — sample data. See desktop view example



