Goldman Sachs Raises Earnings Forecast for Zhongji Innolight as 1.6T Optical Modules Ramp Up
nashnova research
Goldman Sachs resumed coverage of Innolight (中际旭创) on September 7 with a reiterated Buy, implying roughly 225% upside; the core call is that 1.6T optical-module mass production will drive a 78% revenue CAGR from 2026 to 2028 — well above consensus.
What price did Goldman set, and how did the market react?
Goldman set a 12-month target of RMB 2,645 for the A-share and HKD 3,267 for the H-share, both implying roughly 225% upside.
On the day the note dropped, the A-share surged over 10% intraday with turnover topping RMB 30 billion; total market cap climbed back to RMB 1.06 trillion. The H-share jumped as much as 16%.
This means → Goldman sees the current price as reflecting only a fraction of the optical-module upgrade story — and the market voted with real money.
Why is Goldman so much more bullish than consensus?
Goldman forecasts a 78% revenue CAGR for Innolight from 2026 to 2028, driven by the accelerating ramp of 1.6T-and-above optical modules.
Its 2026/27 net-profit estimates sit 25% and 42% above Street consensus, respectively.
In plain terms = most analysts think the optical upgrade is "coming but not that fast." Goldman says "faster and more profitable than you expect." That gap is the core of the current valuation debate.
Why is silicon-photonics leadership the premium driver?
Goldman positions Innolight as a leading global optical-interconnect provider with a significant edge in silicon-photonics (SiPh) modules — modules that use silicon chips instead of traditional materials to carry light signals, cutting cost and easing scale-up.
Two revenue engines ahead: growing SiPh module shipment volumes, and diversification into scale-up (making a single compute cluster bigger) and scale-across (linking multiple clusters) markets.
This means → optical modules are not just "selling components" — they scale with AI compute clusters. The bigger the cluster, the higher the volume and spec requirements.
Could CPO disrupt pluggable optical modules?
CPO — co-packaged optics, embedding the optical module directly inside the chip package instead of plugging it in from outside — is seen as a potential threat to traditional pluggables.
Goldman judges this competitive risk as "manageable" and not a near-term material concern.
Three catalysts Goldman flags for tracking: the 1.6T / 3.2T ramp, spec upgrades driven by next-gen AI platform migration, and incremental revenue from NPO — near-package optics, placing modules closer to the chip while keeping them pluggable.
What is the company itself doing?
Innolight disclosed a RMB 4–8 billion share-buyback plan on August 31; all repurchased shares are earmarked for equity incentives and employee stock-ownership plans.
As of September 4, the company had executed buybacks on four consecutive trading days, spending over RMB 1.3 billion in total.
This reflects management backing its own story with cash — but the real proof point for Goldman's bullish call remains the actual pace of 1.6T module shipments.
市场有风险，内容仅供研究参考，不构成投资建议。