Broadcom’s AI semiconductor revenue rose 221% year over year to $16.7 billion in Q3 FY2026. That is a striking result, but it is not 221% growth in Broadcom’s total revenue—and it does not prove custom AI chips are becoming a bigger business than GPUs. The figures show that demand for custom accelerators and AI networking is growing rapidly at Broadcom; they do not settle the broader market comparison.
What Broadcom’s 221% figure measures
The 221% figure refers to Broadcom’s reported AI semiconductor revenue in its third fiscal quarter of 2026. The company said that revenue was $16.7 billion, up 221% from the same quarter a year earlier and 54% from the preceding quarter. Broadcom’s consolidated revenue for Q3 FY2026 was $29.6 billion, a separate company-wide measure. Broadcom’s Q3 FY2026 financial-results release reported both figures.
Keep the scope in mind: Broadcom’s reported AI semiconductor revenue is not a disclosed custom-chip-only figure. The company attributes growth to demand for custom AI accelerators and AI networking, but its releases do not provide a full revenue split between those categories.
How the reported AI revenue changed by quarter
| Broadcom fiscal quarter | Reported AI revenue | Year-over-year growth | Context |
|---|---|---|---|
| Q4 FY2025 | Not stated in the cited release summary | 74% | Broadcom-reported AI semiconductor revenue growth. Broadcom Q4 FY2025 financial-results release. |
| Q1 FY2026 | $8.4 billion | 106% | Broadcom-reported AI revenue; management cited demand for custom accelerators and AI networking. Broadcom Q1 FY2026 financial-results release. |
| Q2 FY2026 | $10.8 billion | 143% | Broadcom-reported AI semiconductor revenue. Broadcom Q2 FY2026 financial-results release. |
| Q3 FY2026 | $16.7 billion | 221% | Broadcom-reported AI semiconductor revenue; up 54% sequentially. Broadcom Q3 FY2026 financial-results release. |
The series shows accelerating reported growth across the listed quarters. Broadcom’s Q1 release described $8.4 billion as AI revenue, while later releases use the phrase AI semiconductor revenue; the figures are presented here as the company reported them, not as an independently audited market-size series.
#1 Best Overall
- Axial-tech fans now feature a smaller fan hub that facilitates longer blades and a barrier ring that increases downward air pressure
- 2.5-slot design allows for greater build compatibility while maintaining cooling performance
- 0dB technology lets you enjoy light gaming in relative silence
- Dual BIOS switch lets you toggle between Quiet and Performance BIOS profiles
- Dual ball fan bearings last up to twice as long as sleeve bearing designs
Why custom accelerators and networking are part of the story
Broadcom CEO Hock Tan identified demand for both custom AI accelerators and AI networking as drivers. After Q2 FY2026, he said: “Q2 semiconductor revenue from AI of $10.8 billion grew 143% year-over-year, above our forecast, driven by increasing demand for custom AI accelerators and AI networking.” In Q1, he similarly cited “robust demand for custom AI accelerators and AI networking.” These are management explanations of Broadcom’s results, not an independent breakdown of how much revenue came from each product category.
That distinction matters when interpreting the growth. A custom accelerator is designed for particular workloads or a customer’s needs, while networking products connect computing systems and help move data. Broadcom’s reported AI figure includes the company’s AI semiconductor business; the cited releases do not establish that all of it is custom accelerator sales.
Rank #2
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Powered by GeForce RTX 5070 Ti
- Integrated with 16GB GDDR7 256bit memory interface
- PCIe 5.0
- WINDFORCE cooling system
Could custom chips become a bigger business than GPUs?
It is possible that custom accelerators will take a larger role in AI infrastructure, but Broadcom’s quarterly growth alone cannot establish that custom chips will become a bigger business than GPUs. The cited releases report one company’s AI-related revenue and name custom accelerators and networking among the drivers. They do not provide a market-wide revenue comparison between custom chips and GPUs.
Nor do these results show that custom accelerators outperform GPUs or cost less across workloads. A meaningful comparison would need evidence about performance, power use, total system cost, availability, software support and the workloads being served. Those factors can vary by deployment, and the cited company releases do not resolve them.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Rank #3
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Powered by GeForce RTX 5060
- Integrated with 8GB GDDR7 128bit memory interface
- PCIe 5.0
- WINDFORCE cooling system
Guidance is an estimate, not a reported result
Broadcom’s Q2 FY2026 release said management expected Q3 AI semiconductor revenue of $16.0 billion. That was guidance at the time, not a completed-quarter result. The subsequent Q3 release reported $16.7 billion. Keeping forecast and result separate helps prevent a management estimate from being mistaken for realized revenue.
Quick Recap
Best Value
- Axial-tech fans now feature a smaller fan hub that facilitates longer blades and a barrier ring that increases downward air pressure
- Phase-change GPU thermal pad helps ensure optimal heat transfer, lowering GPU temperatures for enhanced performance and reliability
- 2.5-slot design allows for greater build compatibility while maintaining cooling performance
- Dual-ball fan bearings last up to twice as long as standard conventional sleeve bearings designs
- 0dB technology lets you enjoy light gaming in relative silence
Rank #4
- Powered by Radeon RX 9070 XT
- WINDFORCE Cooling System
- Hawk Fan
- Server-grade Thermal Conductive Gel
- RGB Lighting
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
