COHR Stock Outlook 2026: Coherent's 800G and 1.6T Optical Transceiver Bet on AI Infrastructure

Coherent Corp (NYSE: COHR) is the company that sits at the optical interconnect layer of the AI infrastructure stack — the layer between the GPU and the switch. Every Nvidia H100 or GB200 cluster needs thousands of optical transceivers to connect GPUs to top-of-rack switches and data center spine networks. The faster those GPUs get, the more bandwidth the optical layer needs to keep up. COHR makes the optical transceivers, lasers, and compound semiconductor materials that enable this bandwidth. Here is how to think about the 2026 investment case.

---

What Coherent Corp Actually Is: A Photonics Conglomerate

Investors often describe COHR simplistically as “the 800G optical transceiver company” — which captures the most discussed growth driver but misses the scope of what Coherent Corp actually is after the II-VI merger. Understanding the full business helps set appropriate return expectations and risk parameters.

Coherent Corp has three primary business areas:

1. Networking (Data Center and Telecom): This is the 800G/1.6T transceiver business that drives the AI narrative. Includes both short-reach pluggable transceivers for data centers and long-reach coherent optical modules for carrier telecom networks.

2. Lasers: Industrial fiber lasers for metal cutting and welding, diode lasers for materials processing, and specialty laser systems for semiconductor manufacturing and medical applications. This comes primarily from the II-VI heritage and is a $1B+ revenue business in its own right.

3. Materials: Compound semiconductor substrates (InP, GaAs, SiC), optical coatings, and precision optical components. This segment serves both internal manufacturing needs and external customers in defense, medical, and industrial optics.

The Materials and Lasers segments provide genuine diversification against data center demand cycles. When analysts model COHR purely as an AI transceiver play, they sometimes miss that these businesses have their own cycle dynamics — particularly the Lasers segment, which tracks industrial capex.

The AI Infrastructure Optical Stack: Why COHR Exists

How a GPU Cluster Communicates

Imagine 10,000 Nvidia GPUs in a training cluster. Each GPU generates and consumes data continuously during a training run. This data must travel:

• From GPU to top-of-rack switch (< 3 meters, direct attach copper or 400G short-reach optics)
• From rack to rack within a pod (100–500 meters, 400G/800G transceivers)
• Between pods across the data center (500m–2km, 800G/1.6T coherent optics)

The bandwidth requirements scale as a square of the number of GPUs (because every GPU may need to communicate with every other GPU). This is why hyperscaler optical transceiver purchases grow faster than GPU unit purchases.

Where COHR Participates

Network Layer	Speed	COHR Role
Within-rack	400G	Active optical cables, short-reach transceivers
Rack-to-rack (intra-pod)	800G	Pluggable transceivers (primary market)
Inter-pod / spine	1.6T	Emerging, high-ASP opportunity
Telecom long-haul	100G–400G coherent	Established carrier market

---

The II-VI Merger: Building Vertical Integration in Photonics

II-VI Incorporated was already a major supplier of compound semiconductor materials (indium phosphide, gallium arsenide substrates) and industrial lasers when it merged with Coherent in 2022. The combined company now controls its own supply chain from wafer to finished transceiver in a way that fabless photonics companies cannot match.

Why vertical integration matters here:

InP-based components (laser chips inside transceivers) require specialized compound semiconductor manufacturing that is very different from standard CMOS silicon. COHR owns InP epitaxy and fabrication capabilities, giving it both cost control and supply security for the laser chips inside its transceivers. This is a genuine barrier to entry for new competitors.

The merger also combined two complementary customer bases — data center hyperscalers (Coherent’s strength) with industrial laser OEMs (II-VI’s strength) — providing some diversification against any single demand cycle.

---

800G Market: The Current Revenue Driver

800G transceivers are the primary workhorse of current AI cluster builds. Every major hyperscaler (Microsoft Azure, Amazon AWS, Google Cloud, Meta AI infrastructure) has deployed or is deploying 800G networking for GPU fabric interconnects.

Key 800G market dynamics:

• COHR is one of a handful of qualified 800G transceiver suppliers to hyperscalers
• 800G ASPs are significantly higher than 400G, boosting COHR’s revenue per unit
• The transition from 400G to 800G is well underway; the transition from 800G to 1.6T will be the next growth catalyst

Chinese competitors (Innolight, Accelink) are qualified for 400G and are working toward 800G qualification. US geopolitical policy toward Chinese technology supply chains may slow their penetration into leading hyperscaler networks.

---

1.6T: The Next Asymmetric Opportunity

The transition to 1.6T (also called 200G per lane, using 8 lanes for 1.6T total) represents the most important medium-term growth catalyst for COHR.

Why 1.6T is structurally important for COHR:

1. Fewer physical transceivers required per GPU, but each has higher ASP → COHR revenue per unit rises
2. Higher technical complexity → more difficult for Chinese competitors to qualify quickly
3. First-mover advantage in qualification creates sticky revenue

COHR has been publicly discussing 1.6T development alongside its hyperscaler customers. Early design-win confirmation from a Tier-1 hyperscaler would be a strong positive catalyst.

Related: ANET Arista Networks Stock Outlook 2026 →

---

Competitive Landscape: COHR vs. Fabrinet vs. Lumentum

Dimension	COHR	Fabrinet (FN)	Lumentum (LITE)
Business model	Integrated design + mfg	Contract mfg (EMS)	Optical components + laser
800G transceivers	Key supplier	Mfg partner for OEMs	Some supply
Vertical integration	High (InP through module)	None	Medium
Laser business	Strong	None	Strong
AI direct exposure	High	High (through customers)	Medium
Merger risk	Yes (II-VI integration)	None	None

Fabrinet’s business model is complementary rather than purely competitive with COHR — Fabrinet manufactures modules designed by others, including some COHR-designed components. The two companies interact in the supply chain rather than competing purely head-to-head at the transceiver product level.

---

Three Investment Scenarios

Bull Case — Trigger Conditions

1.6T transceiver volume production begins earlier than expected → COHR secures first-mover qualified supplier status with Microsoft and Amazon. AI training cluster build-out accelerates through 2026, driving 800G demand above plan. Telecom carrier spending begins recovering in H2 2026, adding incremental revenue. II-VI integration delivers synergies ahead of schedule.

Monitoring signal: AI data center segment revenue becomes the majority of total revenue in a quarterly filing; 1.6T qualification announcements from hyperscalers.

Base Case

800G demand continues at current elevated levels through 2026. 1.6T qualification progresses, with commercial volumes starting in H2 2026 or 2027. Telecom segment stabilizes. Merger integration on track. Revenue and margin improve gradually.

Bear Case — Trigger Conditions

Hyperscaler AI capex pauses (overspending concerns, model efficiency improvements reduce cluster needs). Chinese 800G competitors qualify with major hyperscalers on cost grounds. Merger integration costs exceed estimates, compressing near-term margins. Telecom carriers cut capex further in a rate-driven economic downturn.

---

Reading COHR’s 10-Q: What Data Center Investors Should Track

Coherent’s quarterly filings require some interpretation because the photonics supply chain terminology is specialized. Here is the practical guide:

Datacom vs. Telecom Revenue Split: COHR separates data center (Datacom) and carrier network (Telecom) revenues. Watch the Datacom percentage of total as the primary growth indicator. A rising Datacom share with growing absolute dollars confirms the AI thesis is playing out.

Product Speed Mix Within Datacom: Not always explicitly broken out, but management will comment on 400G vs. 800G vs. 1.6T mix in conference calls. Rising 800G share with 1.6T mentions indicates the transition is on track.

Gross Margin by Segment: The photonics business carries different margins than the industrial laser business. Rising consolidated gross margin while revenue grows suggests the mix shift toward higher-ASP data center transceivers is working.

Merger Integration Costs: Look for “restructuring charges,” “integration expenses,” and “synergy realization” language. When these charges begin declining as a percentage of revenue, the merger friction is dissipating and normalized profitability becomes visible.

InP vs. SiPh Product Mentions: Any explicit reference to silicon photonics design wins or customer qualifications would be a significant forward-looking indicator, as it represents COHR’s next-generation technology evolution.

---

The InP Supply Chain: COHR’s Compound Semiconductor Moat Explained

The vast majority of investors do not understand why InP (indium phosphide) matters to COHR’s competitive position — and why this is a genuine moat rather than just supplier vertical integration.

What is InP?

Indium phosphide is a compound semiconductor material — not silicon — that enables laser functionality at the wavelengths used for fiber-optic data transmission (1310nm and 1550nm). The physics of light generation in optical transceivers requires III-V semiconductor materials like InP or GaAs. Silicon cannot lase at these wavelengths without assistance.

Why InP manufacturing is hard to replicate

InP epitaxy (growing the layered semiconductor crystal structures) and device fabrication (etching the laser structures from the grown crystal) require specialized equipment, specialized process knowledge, and years of yield optimization. The process technology is fundamentally different from standard CMOS silicon manufacturing that foundries like TSMC excel at.

COHR’s II-VI heritage InP advantage

II-VI Incorporated, before the Coherent merger, had operated InP epitaxy and device fabrication for decades. This accumulated process knowledge — the specific growth parameters, the defect management techniques, the wafer bonding methods — is not transferable quickly even if a competitor builds the physical equipment. It is embedded in the engineering team and process documentation.

This means a new entrant that wants to compete with COHR in 800G+ transceivers cannot simply outsource InP production to TSMC or Samsung. The InP supply chain is a genuine barrier.

---

Silicon Photonics: Opportunity or Threat to COHR?

Investors frequently ask whether silicon photonics — which enables optical functions on standard CMOS silicon wafers — will eventually displace InP-based transceivers and erode COHR’s competitive position.

The technology trade-off

Silicon photonics enables much lower-cost manufacturing in high volumes because it uses existing foundry infrastructure. Intel’s silicon photonics platform and TSMC’s photonics process both target cost-competitive optical integration. The limitation is efficiency — silicon cannot generate light itself, so SiPh still requires an InP laser chip to be integrated (bonded) onto the silicon substrate.

What this means for COHR

COHR is investing in SiPh-based transceiver designs as a complement to its InP designs, not a replacement. For near-term 800G and early 1.6T applications, InP-based designs remain performance-optimal. For future generations beyond 1.6T, SiPh with integrated InP laser chips (co-packaged optics, CPO) becomes increasingly relevant.

COHR’s InP heritage is actually an asset in the CPO world, because the InP laser die that sits inside the CPO module is still a specialized product where COHR has the supply chain. The transition to SiPh does not necessarily disadvantage COHR — it may require new design capability while preserving the InP laser supply role.

---

Hyperscaler Concentration Risk: Modeling a Demand Air Pocket

COHR’s data center transceiver revenue is concentrated among a small number of hyperscaler customers — Microsoft, Amazon, Google, and Meta account for the majority of AI infrastructure spending. This concentration creates an asymmetric downside risk.

Worked hypothetical scenario: Suppose a major hyperscaler pauses its GPU cluster expansion for two quarters due to concerns about AI capex ROI or model efficiency improvements reducing cluster requirements. COHR’s data center transceiver orders from that customer could decline significantly in those two quarters. Because COHR has limited ability to redirect data center-specific transceiver inventory to telecom customers on short notice, the revenue and margin impact would be sharp.

This is the central risk to COHR’s investment thesis, and why the stock is volatile relative to the broader semiconductor sector. Monitoring hyperscaler quarterly earnings calls (Microsoft, Amazon, Google, Meta) for any change in AI infrastructure tone is a first-order investment input for COHR shareholders.

---

Key Metrics to Monitor

1. AI data center transceiver revenue (percentage of total; trajectory)
2. 800G vs. 400G revenue mix shift (800G should be rising)
3. 1.6T design-win announcements or conference call commentary
4. Merger synergy realization (operating expense as % of revenue)
5. Telecom segment revenue trend (stabilization vs. further decline)
6. Hyperscaler capex guidance from Microsoft, Amazon, Google, Meta calls
7. Silicon photonics product development milestones

---

The Industrial Laser Business: Overlooked Diversifier

While AI data center transceivers dominate the investor narrative for COHR, the industrial laser segment provides meaningful revenue and margin diversification that often goes underappreciated.

What industrial lasers do: Coherent’s industrial lasers are used in:

• Metal cutting and welding: Fiber lasers that cut and weld steel and aluminum in automotive manufacturing, shipbuilding, and aerospace fabrication
• Semiconductor manufacturing: Excimer lasers for lithography and annealing; diode pumped solid-state lasers for wafer inspection
• Medical devices: Ophthalmic surgery systems, dermatology lasers, dental tools
• Additive manufacturing: Laser powder bed fusion (LPBF) metal 3D printing

Why this matters for investors: Industrial laser demand is tied to manufacturing capital expenditure cycles — auto OEM tooling investments, semiconductor fab expansions, and aerospace production ramp-ups. This cycle is partially correlated with, but not identical to, the data center AI investment cycle. When one is weak, the other may be stable or growing.

The II-VI side of the COHR merger brought a deep industrial laser portfolio that adds this manufacturing exposure. A pure-play photonics investor focused only on AI transceivers would find the industrial laser diversification either dilutive or valuable depending on macro conditions.

Co-Packaged Optics (CPO): The Next Architecture Shift

Beyond 1.6T transceivers, the photonics industry is watching co-packaged optics (CPO) as the likely architecture for the generation after 1.6T. Understanding CPO helps investors anticipate where COHR’s product roadmap is headed.

What CPO is: Today, optical transceivers are pluggable modules that connect to an Ethernet switch port via a metal cage. In CPO, the optical engine is physically integrated with the switch ASIC on the same package or board. This reduces the copper distance between the switch chip and the optical functions, which reduces power consumption and can increase bandwidth density.

Why CPO matters for COHR: CPO requires the optical die — the InP laser and photonic integrated circuit — to be co-packaged with the switch ASIC. This potentially changes the procurement relationship: instead of selling transceivers to Cisco or Arista, COHR might supply optical engines directly to Broadcom, Marvell, or Intel for integration into their switch ASICs.

COHR’s InP epitaxy capability positions it as a potential CPO optical engine supplier. This is a product category that does not yet generate meaningful revenue but represents the architectural direction of the industry. First-mover design wins in CPO could be a significant medium-term catalyst.

Pre-Investment Checklist for COHR

Before entering a position in COHR:

• Are all four major hyperscalers (Microsoft, Amazon, Google, Meta) guiding for increased AI infrastructure capex?
• Is COHR’s Datacom revenue growing sequentially and year-over-year?
• Has 1.6T qualification been confirmed by any Tier-1 hyperscaler?
• Are merger integration costs declining as a percentage of revenue?
• Is there no evidence of Chinese 800G transceiver qualification at leading hyperscaler accounts?
• Is industrial laser demand stable (providing earnings floor if data center moderates)?

Investment Takeaway

Coherent sits at one of the most AI-leveraged positions in the semiconductor and photonics supply chain. Its 800G transceiver portfolio is already generating meaningful revenue from hyperscaler AI cluster builds, and the 1.6T transition represents a further step-up in ASP that should benefit early qualified suppliers.

The primary risk is a hyperscaler capex air pocket — if AI data center construction pauses for any reason, COHR’s revenue would fall sharply given its concentrated customer base. For investors who believe AI infrastructure spending is durable through 2026–2027, COHR provides a differentiated exposure at the optical interconnect layer that pure GPU-play or networking-switch investments don’t replicate.

The Merger Integration Timeline: What to Expect

The II-VI and Coherent merger closed in early 2022. Large technology mergers of this complexity typically take 2–3 years to fully integrate operations, consolidate manufacturing, and realize the synergies that justified the deal. Investors entering COHR in 2026 are doing so at a point when the most disruptive integration activities should be largely complete.

Integration milestones to track in 2026:

• Manufacturing footprint consolidation: Are there remaining plant closures or consolidations that will create one-time charges? These should be disclosed and declining.
• R&D organization unification: Do COHR and II-VI engineers now operate as a unified team, or are there still organizational silos? Look for increasing cross-platform product announcements as a sign of R&D integration.
• Customer relationship consolidation: Before the merger, some customers bought from both II-VI and Coherent separately. Unified account management should improve cross-sell efficiency.
• Operating leverage: As integration costs fall out of the P&L, the company’s true normalized operating margin should become visible. This can be a re-rating catalyst as investors can model a cleaner earnings structure.

The Telecom Segment: Structural Floor or Drag?

COHR’s telecom revenue stream sometimes draws negative attention from investors focused on the AI growth narrative. Understanding what the telecom business actually is — and is not — clarifies its role in the investment thesis.

What the telecom segment includes: Long-haul coherent optical transceivers for carrier backbone networks, metro coherent systems for regional network infrastructure, and subsea cable optical components. These serve AT&T, Verizon, Deutsche Telekom, NTT, and similar carriers.

Why telecom has been under pressure: After a multi-year period of intense 5G front-haul infrastructure investment, major carriers entered a period of capex digestion in 2023–2025. They had upgraded their access networks substantially and were generating return on that investment before committing to the next upgrade cycle.

Why the telecom floor matters: Even during the capex digestion phase, carriers must maintain and upgrade existing fiber infrastructure. Equipment reaching end of support life must be replaced. The telecom segment provides a revenue floor that prevents total exposure to a single customer category — an insurance policy against hyperscaler capex pauses.

The potential telecom recovery catalyst: Broader AI adoption creates pressure on core network bandwidth (more AI-generated content, more inference traffic). Carriers will eventually need to upgrade their core fiber capacity to handle this traffic growth. When that upgrade cycle begins, COHR’s coherent telecom portfolio is positioned to benefit.

COHR’s Dividend and Capital Return Policy

Coherent Corp has historically not paid a meaningful dividend and has not been a regular share repurchaser, prioritizing R&D and capital investment in manufacturing capacity over shareholder distributions. For income-oriented investors, COHR is not a dividend play. For growth-oriented investors, the absence of dividend obligation means all FCF can be directed toward product development and capacity expansion during the current growth phase.

As the company’s leverage from the II-VI acquisition reduces and FCF generation improves, capital return policy may evolve. Any announcement of a share repurchase program or dividend initiation would signal management’s confidence in FCF durability and could serve as a positive re-rating catalyst.

Positioning COHR in an AI Infrastructure Portfolio

For investors building AI infrastructure exposure across multiple stocks, understanding how COHR relates to other AI supply chain positions clarifies portfolio construction:

AVGO (Broadcom): Designs the custom AI accelerator ASICs (XPUs) and network switch ASICs that hyperscalers use. AVGO’s revenue scales with AI compute silicon deployment, which partially correlates with but is not identical to optical transceiver demand.

ANET (Arista Networks): Sells Ethernet switches that connect GPU clusters. Every Arista switch port needs an optical transceiver — so Arista’s switch shipment data is a leading indicator of optical demand. ANET and COHR are positively correlated.

CRDO (Credo Technology): Makes Active Electrical Cables (AECs) that compete with short-reach optical transceivers at the within-rack level. CRDO and COHR have partially overlapping markets at the shortest-distance layer of the AI network.

COHR’s specific angle: The only public company with both optical transceiver scale AND vertical InP manufacturing AND industrial laser diversification. This combination is not replicated by Fabrinet (contract manufacturer), Lumentum (narrower transceiver focus, no industrial laser), or any pure semiconductor company.

For investors with broad AI infrastructure exposure through NVDA and AVGO, COHR adds the optical layer that neither of those companies directly addresses.

Related: AVGO Broadcom Stock Outlook 2026 → Related: CRDO Credo Technology Stock Outlook 2026 → Related: ANET Arista Networks Stock Outlook 2026 →

---

Disclaimer: This article is for informational purposes only and does not constitute investment advice. Verify all financial data at SEC EDGAR (edgar.sec.gov) and Coherent’s IR site (ir.coherent.com) before making investment decisions.