[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]  Jukan [@Jukanlosreve](/creator/twitter/Jukanlosreve) on x 22.6K followers Created: 2025-06-02 23:39:43 UTC AI ASIC and Networking Chips Broadcom’s dominance in the AI ASIC market is striking, with over XXX chip designs completed across 7nm, 5nm, 3nm, and 2nm process nodes—firmly securing its No.1 position with over XX% market share. In the high-performance switching and routing chip sector, Broadcom also holds a commanding lead, boasting more than XX% market share with its Tomahawk X and Jericho X chipsets. Broadcom is set to begin mass production of its next-generation 3nm Tomahawk X switch chipset in the second half of 2025. The chip supports 102Tbps switching throughput (200Gbps SERDES) and is positioned as the highest-performing Ethernet networking chipset globally, placing Broadcom 1–2 generations ahead of competitors. Meanwhile, Broadcom’s AI ASIC programs are progressing rapidly. Google’s TPUv6 3nm ASIC has entered mass production, featuring two 3nm compute dies and eight HBM3e stacks per package, and is hailed as “the world’s most powerful custom AI XPU.” The project is expected to bring Broadcom over $XX billion in lifetime revenue, with revenue contributions starting in July 2025 and full-scale production in October and into the first half of 2026. Broadcom has completed design for the next-gen TPUv7 3nm, and has already started collaborating with Google on the TPUv8 2nm project, targeting mass production in 2027–2028. With Meta, Broadcom co-designed 7nm and 5nm MTIA AI chips (mainly for inference workloads), and has sampled its next-gen 3nm MTIA chip, which is powerful enough to support both AI training and inference, with mass production expected in 2026. Additionally, Broadcom’s AI ASIC projects with OpenAI and SoftBank/ARM are also steadily progressing, utilizing TSMC’s 3D SoIC chip stacking technology. These projects are expected to tape out in the second half of 2025, with mass production slated for mid to late 2026. These initiatives are set to significantly expand Broadcom’s market opportunity, with projected revenues rising from approximately $XX billion in FY2027 to over $XXX billion in FY2028.  XXXXXX engagements  **Related Topics** [dominance](/topic/dominance) [chips](/topic/chips) [asic](/topic/asic) [coins ai](/topic/coins-ai) [$avgo](/topic/$avgo) [stocks technology](/topic/stocks-technology) [Post Link](https://x.com/Jukanlosreve/status/1929684382824255612)
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
Jukan @Jukanlosreve on x 22.6K followers
Created: 2025-06-02 23:39:43 UTC
AI ASIC and Networking Chips
Broadcom’s dominance in the AI ASIC market is striking, with over XXX chip designs completed across 7nm, 5nm, 3nm, and 2nm process nodes—firmly securing its No.1 position with over XX% market share. In the high-performance switching and routing chip sector, Broadcom also holds a commanding lead, boasting more than XX% market share with its Tomahawk X and Jericho X chipsets.
Broadcom is set to begin mass production of its next-generation 3nm Tomahawk X switch chipset in the second half of 2025. The chip supports 102Tbps switching throughput (200Gbps SERDES) and is positioned as the highest-performing Ethernet networking chipset globally, placing Broadcom 1–2 generations ahead of competitors.
Meanwhile, Broadcom’s AI ASIC programs are progressing rapidly. Google’s TPUv6 3nm ASIC has entered mass production, featuring two 3nm compute dies and eight HBM3e stacks per package, and is hailed as “the world’s most powerful custom AI XPU.” The project is expected to bring Broadcom over $XX billion in lifetime revenue, with revenue contributions starting in July 2025 and full-scale production in October and into the first half of 2026.
Broadcom has completed design for the next-gen TPUv7 3nm, and has already started collaborating with Google on the TPUv8 2nm project, targeting mass production in 2027–2028.
With Meta, Broadcom co-designed 7nm and 5nm MTIA AI chips (mainly for inference workloads), and has sampled its next-gen 3nm MTIA chip, which is powerful enough to support both AI training and inference, with mass production expected in 2026.
Additionally, Broadcom’s AI ASIC projects with OpenAI and SoftBank/ARM are also steadily progressing, utilizing TSMC’s 3D SoIC chip stacking technology. These projects are expected to tape out in the second half of 2025, with mass production slated for mid to late 2026.
These initiatives are set to significantly expand Broadcom’s market opportunity, with projected revenues rising from approximately $XX billion in FY2027 to over $XXX billion in FY2028.
XXXXXX engagements
Related Topics dominance chips asic coins ai $avgo stocks technology