Taiwan ODM/Brands AI server/PC supply chain: AI server rack ramps and new smartphones are expected to lift Taiwan ODM/brand revenues through late 2026

Overview

This three-month revenue preview assesses 10 Taiwan ODM and brand companies across the AI-server, PC and consumer-electronics supply chain. Goldman Sachs expects AI-server rack ramps and new smartphone models to sustain strong year-on-year growth into late 2026, but sees company-specific differences from PC weakness, component availability, valuation and execution.

Core views

Goldman Sachs expects the 10 covered companies’ average revenue to rise +7%/+2%/+4% month on month in September/October/November 2026 and +49%/+57%/+52% year on year. The firm attributes the outlook to next-generation rack-level AI servers, higher specifications and ASPs, expanding ASIC AI-server adoption, PC product-mix upgrades, new flagship smartphones, and higher memory costs lifting selling prices. It is positive on Apple’s supply chain as new form factors—including foldable phones in 2026 and iPhone 20 in 2027—support demand, as well as AI-server components and ASIC-server penetration. For Hon Hai, Goldman Sachs forecasts 3Q26 and 4Q26 revenue growth of +14% and +23% QoQ, respectively. It expects early delivery of next-generation AI-server racks and CPO switches, around two to three months ahead of peers, to support revenue from September, followed by GPU racks and LPU-powered racks from December. New premium and foldable smartphone models are a further driver despite the firm lowering global smartphone volume estimates by 1m/8m/0.2m units to 1.1bn/1.2bn/1.2bn in 2026-28E. Hon Hai remains Buy-rated with a NT$400 12-month target based on 17.9x 2027E P/E. The report expects AI-server transitions to produce mixed near-term patterns. Quanta's PC shipments fell 30%/37% YoY in July/August, and Goldman Sachs forecasts only +2% QoQ revenue growth in both 3Q26 and 4Q26 before a December acceleration as new AI racks ramp. It estimates Quanta capex at NT$44bn/NT$32bn in 2026/27E versus NT$15bn in 2025, citing AI and general-server demand and capacity diversification. The firm retains Neutral with a NT$299 target at 10.7x 2027E P/E. Gigabyte is projected to accelerate to 81%/79% YoY revenue growth in 3Q/4Q26 as AI spending, customer deployment and backlog visibility improve; its 2026 capex is expected to rise 143% YoY to NT$2.8bn. Goldman Sachs nevertheless retains Neutral and a NT$487 target. AVC is expected to deliver +18%/+27% QoQ revenue growth in 3Q/4Q26 as GPU and ASIC server-rack launches raise liquid-cooling adoption, a major ASIC customer ramps new models, and customer expansion supports 4Q26. Goldman Sachs forecasts AI server-rack shipments of 55k/105k/163k in 2026-28E and AVC net-income CAGR of +39% over that period. It maintains Buy and a NT$4,513 target based on 30.3x 2027E P/E. Wiwynn's August revenue beat Goldman Sachs estimates by 27%, supported by general-server demand from leading US CSPs and new ASIC AI-server ramps. The firm forecasts +46%/+22% QoQ growth in 3Q/4Q26, sees GPU server racks entering mass production in 4Q26, and estimates NT$44bn 2026E capex; Buy is retained with a NT$3,777 target at 21.5x 2027E P/E. Wistron's August revenue rose 166% YoY to NT$460bn, 48% above Goldman Sachs estimates, driven by rack-level AI-server shipments despite sluggish PC demand. The firm raises 2026-28E revenue estimates by 11%/7%/6% and net-income estimates by 2%/3%/4%, while lowering gross-margin assumptions because rack-level servers carry lower margins. It raises the 12-month target to NT$295 from NT$281, using 13.2x 2027E P/E versus 13.1x previously, and maintains Buy. Inventec's August revenue was +36% YoY and broadly in line, but component shortages constrained sequential performance. Goldman Sachs raises 2026-28E revenue by 8%/10%/6% and net income by 5%/4%/2%, raises its target to NT$61 from NT$56, and maintains Neutral. For consumer-PC-exposed names, Goldman Sachs expects growth but remains cautious on demand and valuation. ASUS is expected to grow revenue 39%/27% YoY in 3Q/4Q26, aided by higher-end PC mix and AI-server ramps, but 4Q revenue is forecast to decline 7% QoQ following demand pull-in; Neutral is retained with a NT$1,115 target. Compal's August revenue missed estimates by 20% as PC shipments weakened, but AI-server ramps and a move from motherboard to system assembly support expected +1%/+16% QoQ revenue growth in 3Q/4Q26; Goldman Sachs retains Neutral and NT$46.1. Pegatron is projected to grow revenue +18%/+25% YoY in 3Q/4Q26 on AI servers, smartphone launches and seasonal broadband, IoT and console demand, but remains Sell because valuation is viewed as stretched; its NT$76.9 target uses 10.8x 2027E P/E.

Analysis framework

Goldman Sachs reviews August reported revenue against its estimates, projects monthly and quarterly revenue through December, and links each outlook to AI-server platform ramps, ASPs, PC demand, smartphone launches, capacity investment and supply constraints. It values individual companies mainly using forward P/E multiples calibrated against peer relationships between earnings growth, operating margins where applicable, and valuation multiples.

Methodology notes

• Valuation methodsP/E and PEG Valuation

  Forward P/E valuation calibrated to peer relationships between earnings growth, margins and valuation multiples.

  Goldman Sachs applies target P/E multiples to 2027E EPS for the covered companies, generally referencing peer correlations between forward earnings growth and P/E; Hon Hai also uses a PEG-and-margin comparison.

• Industry AnalysisSupply-demand framework

  AI-server, PC, smartphone and component supply-demand analysis.

  The report connects customer demand, product ramps, component availability, capacity investment and pricing to monthly and quarterly revenue forecasts.

• Industry AnalysisVolume-price decomposition

  Revenue effects from shipment volumes and higher ASPs.

  The analysis distinguishes shipment ramps from price effects, particularly higher ASPs for rack-level AI servers and memory-cost-driven selling-price increases.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Hon Hai (2317.TW)

  Early next-generation AI-server rack and CPO-switch delivery plus new smartphones support growth.

  Strengths

  Expected first mover in latest AI-server racks; foldable and premium smartphone exposure.

  Comparison

  Expected to deliver latest model around two to three months ahead of peers.

  Risks

  Slower AI-server ramp, weaker EV total-solution performance, slower global capacity expansion, or stronger EMS competition.

• Quanta (2382.TW)

  AI-server ramps and high capex offset slowing PC shipments.

  Strengths

  AI and general-server demand; capacity diversification.

  Weaknesses

  PC shipment slowdown and AI-server model transition.

  Comparison

  Fair valuation at 10.7x 2027E P/E.

  Risks

  PC recovery, AI-server ramp and general-server demand may differ from expectations.

• AVC (3017.TW)

  Liquid-cooling adoption and ASIC/GPU AI-server ramps drive growth.

  Strengths

  Air- and liquid-cooling total-solution position.

  Comparison

  Target P/E derived from technology-hardware peer growth and valuation correlation.

  Risks

  Slower liquid-cooling adoption, weaker AI-server demand, competition or slower general-server recovery.

• Wiwynn (6669.TW)

  ASIC, GPU and general-server demand support revenue and capacity expansion.

  Strengths

  Strong August beat, expanding AI-server platforms and geographic production diversification.

  Comparison

  Target multiple references PC/server peers' EPS-growth versus P/E relationship.

  Risks

  Weaker AI-server demand, slower general-server recovery or tougher server-ODM competition.

• Wistron (3231.TW)

  Rack-level AI-server shipments drive earnings upgrades and a higher target price.

  Strengths

  Market-leading AI-server ODM position and planned US/Vietnam capacity investment.

  Weaknesses

  Higher rack-server mix lowers gross-margin assumptions; PC demand remains weak.

  Comparison

  Target multiple is between historical average and average plus one standard deviation forward P/E.

  Risks

  Weak PC recovery, slower AI-server ramp or weaker general-server demand.

• Gigabyte (2376.TW)

  AI-server deployment, broader customers and capacity investment drive stronger 2H26 growth.

  Strengths

  Improving backlog visibility and production diversification.

  Comparison

  Neutral valuation at 14.1x 2027E P/E.

  Risks

  PC-market recovery, gaming-PC growth and AI-server ramp may differ from expectations.

• ASUS (2357.TW)

  AI-server expansion and higher-end PC mix support growth.

  Strengths

  AI-server expansion and PC mix upgrade.

  Weaknesses

  Demand pull-in and AI-server model transition are expected to cause a 4Q26 sequential decline.

  Comparison

  Valuation uses brand-maker P/E, earnings-growth and operating-margin relationships.

  Risks

  PC, AI/gaming-PC and AI-server growth may differ from expectations.

• Compal (2324.TW)

  AI-server systems expansion is expected to support late-2026 growth.

  Strengths

  Expansion from L6 motherboards to L10/L11 systems and global capacity diversification.

  Weaknesses

  August revenue miss, weak PC shipments and consumer demand exposed to raw-material costs.

  Comparison

  Neutral valuation at 14.7x 2027E P/E.

  Risks

  PC recovery, AI-server ramp and tablet demand may differ from expectations.

• Inventec (2356.TW)

  AI-server expansion supports estimates, while component shortages weigh on near-term growth.

  Strengths

  Rising AI-server ASP assumptions and revised earnings outlook.

  Weaknesses

  Tight memory, PCB and CPU supply; sluggish PC demand.

  Comparison

  Target P/E is based on local and global PC/server peer earnings-growth correlations.

  Risks

  PC recovery, AI-server ramp and general-server demand may differ from expectations.

• Pegatron (4938.TW)

  AI servers and new smartphone launches support revenue, but Goldman Sachs retains Sell on valuation.

  Strengths

  Expansion into AI servers and automotive electronics; seasonal broadband, IoT and console demand.

  Weaknesses

  High memory costs cap PC and smartphone demand; August revenue missed estimates.

  Comparison

  Sell rating reflects what Goldman Sachs describes as stretched valuation.

  Risks

  Stronger smartphone recovery, faster AI-server ramp or stronger consumer-electronics demand.

Key data

• Average 10-company revenue growth, September/October/November 2026+7%/+2%/+4% MoM; +49%/+57%/+52% YoYGoldman Sachs forecast for the covered supply chain.
• Wistron August 2026 revenueNT$460bn, +166% YoY48% above Goldman Sachs estimate.
• Wistron 2026-28E net-income revision+2%/+3%/+4%Driven mainly by higher rack-level AI-server revenue.
• Inventec 2026-28E net-income revision+5%/+4%/+2%Driven mainly by higher revenue and AI-server ASP assumptions.
• AVC 2026-28E net-income CAGR+39%Attributed to AI spending, liquid-cooling penetration and total-solution positioning.
• Wiwynn August revenue versus estimate+27%Supported by general-server demand and new ASIC AI-server ramp-up.

Impact & implications

The report argues that the supply chain is shifting toward higher-value rack-level AI-server systems, liquid cooling and ASIC platforms, supporting strong revenue growth and capacity expansion. However, the benefit is uneven: PC demand pull-in, high memory costs, model transitions, lower server margins and component shortages can constrain particular companies or periods.

Risks

• AI-server demand or platform ramps could be slower than expected.
• High memory costs, weak PC demand after first-half pull-in and component shortages could constrain growth.
• Competitive pressure in server ODM, cooling and consumer-electronics EMS could intensify.
• Rack-level AI-server mix may dilute gross margins for certain ODMs.
• Capacity expansion and production-site diversification may not proceed as expected.

What to watch

• Monthly September-to-December revenue progress against the forecast +7%/+2%/+4% average MoM growth path.
• The ramp timing for next-generation GPU, ASIC and LPU AI-server racks and associated liquid-cooling adoption.
• PC shipment recovery, memory costs and the availability of memory, PCB and CPU components.
• New flagship and foldable smartphone launches and their effect on ODM demand.
• Company capex and capacity-expansion execution, particularly in the US, Mexico, Malaysia and Vietnam.