๐Ÿ“ก Research Board โ€” Created by GWY

Daily & weekly automated equity research โ€” semis / AI / tech
SG --:--:-- NY (ET) --:--:-- ๐Ÿ“… -- Dark Mode
๐Ÿ“Š View earnings presentation
๐Ÿ“„ Source: Investing.com
โšก Q/Q Change Highlights
  • Revenue $804.2M vs ~$751M Q4 (+7% QoQ, +26% YoY) โ€” record; beat $781.6M cons; EPS $5.10 vs $4.90 est
  • Comms +33% sequentially โ€” power for optical modules + switches
  • Enterprise data growth floor raised to +85% YoY (from 50%) on strong, extended ordering
  • Capacity goal raised: past $4B plan โ†’ $6B target in the near future
  • GM 55.5% (flat, low end of model); 800V sampling; DDR5 high-speed interface initial sampling (not a 2026 revenue contributor)

๐ŸŽ™๏ธ MPWR โ€” Apr 30, 2026

๐Ÿ“„ Original Transcript

Monolithic Power Systems (MPWR) Q1 2026 Earnings Call Transcript

Date: April 30, 2026 | Source: investing.com / company press release

---

Arthur Lee (IR): Earlier today, MPS released a written commentary on our results of operations for the first quarter ended March 31, 2026. This document can be found on our website. Our statements are made as of today, and we assume no obligation to update this information. I would like to turn the call over to Tony.

Tony Balow (CFO): Thanks, Arthur. Good afternoon, and welcome to our Q1 2026 earnings call. In Q1, MPS achieved record quarterly revenue of $804 million, 7% higher than the fourth quarter of 2025 and 26% higher than the first quarter of 2025. Our quarterly performance was a result of our continued innovation, our consistent execution, and the resilience of our diversified market strategy. Let me call out a few highlights from the quarter. Our communications end market grew 33% sequentially on the strength of our power solutions for optical modules and switches. The pipeline for our automotive and enterprise data end markets, including server, continue to accelerate as we won multiple new projects across customers and regions.

MPS continued to grow our capacity past our original $4 billion plan with a new goal of reaching $6 billion in the near future. We continue to adjust to the fluid geopolitical and macroeconomic environment. Our diversified market strategy remains unchanged. Operator, you may now open the webinar for questions.

Ross Seymore (Deutsche Bank): I just wanted to dig a little bit into the enterprise data side of things.

Tony Balow: Ross, if you recall, even last year in 2025, we had talked about CPU being a tailwind, and we continue to see that here in 2026. If you look across enterprise data, it's increasingly hard to differentiate between AI solutions and CPU, but in general, all the growth drivers are intact. We're ramping new customers, we've been ramping existing customers, and we continue to see the transition to modules. Plain server has been a tailwind, and we think it will continue to be so.

Michael Hsing (CEO & Founder): The notebooks, we don't really care this quarter or next quarter. As long as we develop the best solutions, our power densities, our customers' ease of use and the design wins that we have, the revenue will run.

William Stein (Truist): You noted passing the $4 billion capacity target and working toward $6 billion. Can you update us on the geographic placement and technology?

Michael Hsing: We are still around the 60 nanometers, and maybe we'll go down to 40-45 nanometers as power density stays a market trend and power increases. For the $6 billion goal for manufacturing pipelines, we clearly see our near futures โ€” a lot more activities, a lot more potential. All these design wins are imminent and they'll turn into our revenues.

Tony Balow: On the $4 billion of capacity, we talked about that being very geographically diverse, both inside and outside of China. Our strategy is to maintain that supply chain diversity, and we'll continue to try to have that balance going forward.

Joshua Buchalter (TD Cowen): Can you help us with the enterprise data growth floor?

Tony Balow: We tend to be fairly conservative, waiting for the backlog to be in place. Late last year, we talked about 30%-40% growth year-over-year. In the last call, we raised that to a 50% floor. The strong ordering patterns that we saw start last year have continued through Q1. At this point in time, we're comfortable raising that floor up to around 85% year-over-year growth. That will certainly be one of the drivers of growth for the year for MPS.

Rick Schafer (Oppenheimer): The top 4 CSPs are now over $700 billion in CapEx. Are you able to capture all of that upside? Is anything curbing your supply?

Michael Hsing: I think that's exactly right. A few quarters ago, in AI GPU power, at any given time, the performance, manufacturing capabilities and reliability will remain with only a few players. After a couple of years, it's very clear MPS is one of the players. We continue to do well, especially on the power density side. We provide total monolithic power solutions โ€” a single piece of silicon versus competitors using multiple pieces. Yet we don't want to be the dominant supplier; our goal is diversified growth.

Michael Hsing (on physical AI): We see this year the volume is still low, but it can move the needle slightly. We're at the very beginning โ€” it's very difficult to predict. We clearly benefit from it. We're trying to run the typical MPS playbook โ€” engage broadly and win all the designs we can. We can't control when customers ramp, but we can control winning the sockets.

Quinn Bolton (Needham): On optical modules โ€” 800G modules are more than doubling in 2026. When should we see material revenue from the DDR5 high-speed interface sampling?

Michael Hsing: We see a lot of activity and demand for high power density products, especially modules. The power density of the module in a very confined area, and the data rate keeps increasing. With opticals or other formats, the power will keep increasing. On high-speed DDR5, our engineers created it โ€” we have pretty good positions in PMIC in memory, and we introduce timing drivers, timing control, temperature sensors. The revenue usually we don't talk about a product until we sample those products.

Tony Balow: I wouldn't really have that as a contributor to 2026 revenue. We're highlighting it as we continue to expand our footprint in that market.

Tore Svanberg (Stifel): On the move to 2,000-watt GPUs and 800V โ€” what makes MPS unique?

Michael Hsing: MPS is focused on monolithic solutions โ€” the most cost-effective integrations, the ability to integrate or disintegrate, put it in modules. That's a huge advantage. We invested in a module development unit for other segments starting in 2016. We know how we test and qualify these devices โ€” we develop our own test systems and reliability systems, fully automated, based on MPS motion products. That's a huge advantage.

Michael Hsing (on 800V): We're sampling and co-developing systems with our customers and our customers' customers. Our products for that application are ready. We also have 800 volts going to higher voltages โ€” other segments have to be developed a lot more, more efficient power conversions. MPS will play in those segments.

Gary Mobley (Loup Capital): On the comms business โ€” can you give color on content?

Tony Balow: We're going to stop short of giving a dollar content. In optical modules, we have a module-in-the-module. We obviously look at more of that than a discrete device. If you talk about switches, you have a different number of trays โ€” switches, NIC cards, other things that all require power. We're not going to talk about specific content layers, especially for specific customers.

Joseph Quattrocchi (Wells Fargo): Puts and takes on the gross margin guide?

Michael Hsing: The margins in the last couple of quarters are on the low end. Although it's in our models, it's on the low end. We still improve the yields on the modules. We don't have much of a headwind. We're moving up, but I don't want to give you false hope that we're going to jump very high.

Tony Balow: Historically, we've been very consistent delivering to our gross margin guide. For the last 4 quarters, we've been flat at 55.5%, at the low end of our gross margin model (55% to upper 50s%). For Q2, we did have the confidence to increase incrementally our gross margins because we've gotten better visibility to our backlog. We do see some strong headwinds potentially in the second half, so we're remaining cautious for the guide in the second half.

Kelsey Chia (Citi): Some peers are using GaN for higher-voltage step-down. Your thoughts?

Michael Hsing: I said I didn't believe in GaN, and I still don't believe that for high power โ€” it still has to be proven in the market segment. We have deep know-how to use silicon carbide. Unlike other companies, we don't sell silicon carbide FETs โ€” we always integrate it into our modules. On supply, we don't play a passive role. When customers tell us to pull in too late, we've actively, preemptively built inventory. Our product life cycle is very long. The 85% enterprise data floor is not because of any constraints in the supply chain โ€” that's not an issue right now.

Tony Balow: Nothing about our outlook or anything we said about the enterprise data floor is because we see any constraints in the supply chain. We've continuously stayed ahead of it.

๐Ÿ“ Summary

MPWR (Monolithic Power Systems) โ€” Q1 2026 (Apr 30, 2026). Record quarter: rev $804.2M (+7% QoQ, +26% YoY, beat); comms +33% seq; enterprise data growth floor raised to +85% YoY.

Results

  • Revenue: $804.2M record (+7% QoQ, +26% YoY; beat $781.63M); EPS $5.10 non-GAAP (beat $4.90); GM 55.5% (flat, low end of 55-upper-50s model)
  • Comms +33% seq (optical modules + switches); all growth drivers in enterprise data intact (AI + CPU + modules + new/ramping customers)
  • Enterprise data floor raised to ~85% YoY growth (from 30-40% โ†’ 50% โ†’ 85% progression)
  • Capacity: passed $4B plan, new ~$6B goal; geographically balanced (China/non-China); monolithic single-die power solutions = density advantage
  • 800V AC-DC sampling (SiC-based; own GaN for low-voltage); DDR5 high-speed interface sampling (memory PMIC + timing drivers); robotics/building-automation TAM ~$40-50B
  • Auto: flat 1H, ramp 2H; diversified growth goal (won't be dominant supplier)

Guidance

  • Q2 (Jun qtr): continued record trajectory; GM incrementally up (better backlog visibility) but cautious on 2H headwinds
  • FY26: enterprise data +85% floor; capacity ~$6B; auto ramp 2H
  • No specific revenue dollar guidance disclosed (management remains conservative)

Capex

  • Capacity expansion to ~$6B revenue capability (past $4B plan); balanced China/non-China; increasingly investing in back-end/module assembly (higher complexity); no specific capex dollar guide

Key Q&A

  • Q (Seymore, DB): Enterprise data drivers?
    A: CPU tailwind continues; increasingly hard to differentiate AI vs CPU; ramping new + existing customers; transition to modules; plain server a tailwind
  • Q (Stein, Truist): Capacity geography/tech?
    A: ~60nm, going to 40-45nm; $6B capacity goal; very geographically diverse (inside/outside China) to maintain supply-chain diversity
  • Q (Buchalter, TD): ED growth floor?
    A: Conservative by design (backlog first); floor raised 50% โ†’ 85% YoY on continued strong ordering through Q1
  • Q (Schafer, Oppenheimer): Capture all CSP capex upside?
    A: MPS among few players with performance/reliability at scale; single-die monolithic advantage; goal is diversified growth, not dominance
  • Q (Bolton, Needham): DDR5 high-speed interface timing?
    A: Not a 2026 revenue contributor; expanding footprint in memory (PMIC + timing)
  • Q (Svanberg, Stifel): 2,000W GPU / 800V differentiation?
    A: Monolithic integration + own automated test/reliability systems (since 2016); 800V ready, based on SiC; MPS will play in higher-voltage segments
  • Q (Mobley, Loup): Comms content?
    A: Module-in-module in optical; switches/NIC/trays all require power; no specific content disclosure
  • Q (Quattrocchi, Wells): GM puts/takes?
    A: Flat 55.5% at low end for 4 qtrs; Q2 incrementally higher (backlog visibility); cautious on 2H headwinds
  • Q (Chia, Citi): GaN vs SiC?
    A: Doesn't believe GaN for high power; deep SiC know-how integrated into modules; 85% ED floor NOT supply-constrained

Notes

  • Record Q1 with enterprise data floor raised to +85% โ€” the 50%โ†’85% progression shows accelerating AI/server power demand; comms +33% seq adds an optics-driven second engine
  • Capacity (~$6B goal) + module strategy + 800V/DDR5/robotics optionality = multi-year growth levers; GM deliberately held at low end of model to drive growth
  • Watch: 2H GM headwinds (called out), auto 2H ramp, 800V data-center architecture timing, and whether the 85% ED floor moves higher