Competitors
Coherent operates across four distinct competitive layers, each with different dynamics, margin economics, and competitive intensity:
- Merchant InP EML / VCSEL source-laser supply — a structural duopoly with Lumentum
- Datacom transceiver modules — competition with Innolight, Eoptolink, plus a long tail of Asian module-makers
- SiC substrate — competition with Wolfspeed, STMicroelectronics, plus Chinese suppliers (TankeBlue, SICC)
- Industrial fiber lasers — competition with IPG Photonics, Trumpf, and a tail of Asian players
There is also a fifth, longer-cycle competitive layer: Marvell as DSP partner today but with Marvell-Polariton in-house silicon-photonics platform as a multi-year alternative pathway that could displace some merchant-photonics demand.
Tier 1 — Merchant InP EML / VCSEL duopoly with Lumentum
Lumentum Holdings (NASDAQ: LITE; CIK 0001633978) is Coherent’s structural duopoly partner in the merchant InP source-laser supply business. Together they account for the overwhelming majority of merchant InP EML and VCSEL supply globally. The duopoly was forged through two parallel M&A waves:
| Year | M&A event | Effect on InP/VCSEL merchant supply |
|---|---|---|
| 2018 | Lumentum acquires Oclaro ($1.7B) | Consolidates Towcester UK InP fab and tunable lasers into Lumentum |
| 2019 | II-VI acquires Finisar ($3.2B) | Brings Finisar Sherman TX InP/VCSEL fabs into II-VI |
| 2022 | II-VI acquires Coherent Inc. ($7.01B) | Combined entity rebrands as Coherent Corp Sept 1 2022 |
| 2022 | Lumentum acquires NeoPhotonics ($918M) | Adds advanced InP and silicon-photonics components |
| 2023 | Lumentum acquires Cloud Light ($750M) | Adds module-assembly capability — narrows the vertical-integration gap |
Post-2022 stable structure: two near-equal merchant InP-laser suppliers, both vertically integrated through finished-module assembly, both with US fab capability, both targeting the AI-photonics demand wave. NVIDIA’s parallel $2B equity investments in both companies on March 2, 2026 is the explicit recognition by the largest end-customer that the duopoly is structurally important — NVIDIA is reinforcing both legs rather than picking a winner. See nvidia partnership.
Coherent’s structural differences vs Lumentum
| Dimension | Coherent | Lumentum |
|---|---|---|
| InP wafer size in production | 6-inch / 150mm at Sherman TX since pre-2024 | 4-inch / 100mm; transitioning at Greensboro NC |
| Vertical integration | Full vertical: InP fab + SiPh + assembly + materials | InP fab + SiPh + Cloud Light assembly |
| Materials portfolio | Broad: SiC + ZnSe + GaAs + ICO + rare-earth | Narrower: optical-component focused |
| 3D-sensing VCSEL exposure | Direct — Sherman TX VCSEL line (Apple Face ID 2nd source via Finisar 2019) | Direct — Apple Face ID original primary supplier |
| Industrial fiber laser exposure | Yes — legacy Coherent Inc. franchise | None |
| NVIDIA equity instrument (Mar 2 2026) | Common stock ($256.80 × 7,788,161 = $2B) | Series A Convertible Preferred ($695.31 × 2,876,415 = $2B) |
| Scale | Larger by revenue, market cap, and fab footprint | Smaller; pure-play optical |
Coherent’s 6-inch InP wafer migration is the single most-cited competitive differentiator: a 6-inch wafer yields ~2.25× the laser chips per fab floor area vs a 4-inch wafer at lower per-chip cost. This is a structural cost-of-production lead that Lumentum is closing via the Greensboro NC fab announced in March 2026.
The Coherent-Lumentum relationship is dual-natured: in some product categories Lumentum is a supplier to Coherent (NeoPhotonics-heritage components flow into Coherent transceivers), and in others Coherent is a supplier to Lumentum (Sherman TX components into Lumentum modules). They are not pure adversaries — the merchant-laser franchise benefits both from supply discipline.
Tier 2 — Datacom transceiver-module competition
The 800G / 1.6T pluggable transceiver market is dominated by a handful of large module-builders, with Chinese players (Innolight + Eoptolink) controlling the largest combined market share for NVIDIA-bound modules per industry trade press.
NVIDIA-bound 800G optical module supplier shares (◐ industry-attributed via TrendForce / Cignal AI / IP-Fiber)
| Module supplier | Estimated NVIDIA-bound 800G share | Notes |
|---|---|---|
| Innolight (China — SHE: 300308) | ~35% | 2024 revenue RMB 23.86B (~$3.3B); the global #1 optical module-maker |
| Eoptolink (China — SZ: 300502) | ~25% | 2025 H1 revenue RMB 10.4B (+283% YoY); rapidly scaling |
| Coherent / Finisar legacy | Mid-teens | Vertically integrated direct-from-laser-fab |
| Lumentum / Cloud Light | Low-double-digit | Post-Cloud Light scaling |
| Others (Accelink, AOI, Source Photonics, etc.) | Tail | Various Asian module-builders |
Note: the InP source-laser inside the Chinese module-makers’ product still mostly comes from Lumentum or Coherent. The “Innolight + Eoptolink dominate 60%” framing refers to module-assembly market share, not laser-chip share. The Chinese module-makers buy InP EMLs from Lumentum and Coherent and assemble them with Marvell / Broadcom / Macom DSPs into finished pluggable transceivers.
The competitive structure is therefore complementary rather than zero-sum at the laser-chip layer: more Innolight/Eoptolink module shipments = more InP EML chip demand at Coherent and Lumentum. The squeeze is on module margins, not on laser-chip margins. Coherent’s vertical-integration into module assembly (Finisar heritage) is the hedge against margin erosion at the module layer.
Cisco-Acacia and other Western module players
| Vendor | Position |
|---|---|
| Cisco-Acacia | Coherent-pluggable transceivers (400ZR, 800ZR); competes with Coherent at the module level for telecom pluggable applications |
| Marvell | DSP supplier (not module-maker today) — but Marvell-Polariton EO-polymer is a forward alternative |
| Broadcom | DSP + VCSEL legacy (limited datacom-transceiver business at module level) |
Tier 3 — SiC substrate competition
Silicon carbide substrates are a non-photonics franchise but materially important to Coherent’s revenue mix and to its strategic narrative as a US-domestic compound-semiconductor leader.
| Vendor | Status | Notes |
|---|---|---|
| Wolfspeed (NYSE: WOLF) | Dominant US producer of 200mm SiC substrates | Multi-year capacity expansion at Mohawk Valley NY fab; financial distress through 2024-2025 |
| STMicroelectronics (NYSE: STM) | Vertically integrated SiC producer + power-device manufacturer | Catania Italy fab; integrated supply for STM’s own SiC MOSFETs |
| Coherent Corp (legacy II-VI) | Long-standing 150mm/200mm SiC substrate producer | Easton PA fab; merchant-supply oriented |
| Onsemi (NYSE: ON) | SiC substrate via GT Advanced Technologies (acquired 2021) | Vertical for Onsemi’s own SiC MOSFETs |
| TankeBlue (China, private) | Rapidly scaling Chinese SiC substrate producer | Beijing-based; aggressive capacity ramp |
| SICC (China, private) | Major Chinese SiC substrate producer | Shandong-based; aggressive capacity ramp |
| ROHM (TSE: 6963) | Japanese vertically integrated SiC producer | SiCrystal subsidiary in Germany |
The SiC market dynamic is margin-pressured: TankeBlue and SICC are scaling Chinese capacity rapidly, EV demand growth has slowed below pre-2024 expectations, and Wolfspeed’s financial distress has triggered industry-wide pricing pressure. Coherent’s SiC franchise is profitable but the segment growth thesis has weakened materially through 2025–2026. ◐ — see sic substrates for detail.
Tier 4 — Industrial fiber laser competition
Industrial fiber lasers (cutting/welding/additive manufacturing/marking) are the legacy Coherent Inc. franchise inherited via the 2022 merger. The competitive set is mature:
| Vendor | Status | Notes |
|---|---|---|
| IPG Photonics (NASDAQ: IPGP) | Long-standing global leader in high-power industrial fiber lasers | Vertically integrated from active-fiber pump-laser through finished system |
| Trumpf (private, German) | Largest privately-held industrial-laser company globally | Diversified across laser sources + machine tools |
| Coherent Corp (legacy Coherent Inc.) | Major industrial-laser supplier across cutting/welding/additive/life-sciences | Combined post-2022 with II-VI’s optics franchise |
| Han’s Laser (SHE: 002008) | Largest Chinese industrial laser company | Aggressive pricing in lower power tiers |
| Maxphotonics, Raycus, JPT (China) | Chinese mid-tier industrial fiber laser players | Aggressive pricing |
The industrial-fiber-laser market is structurally low-growth with pricing pressure from Chinese competitors. Coherent’s exposure here was the rationale for diversifying via the 2022 acquisition — but the segment is not the AI-photonics growth story. ✓
Tier 5 — DSP partners and the Marvell-Polariton long-cycle alternative
Marvell is Coherent’s most important DSP supplier — Marvell DSPs (the post-Inphi-acquisition coherent-DSP line) are integrated into Coherent’s pluggable transceivers. ✓ At the module level Marvell is not a competitor.
But Marvell acquired Polariton Technologies in 2024-2025 — a Swiss/European silicon-photonics + electro-optic-polymer company — and is internally building a silicon-photonics + EO-polymer integrated platform. The forward question for Coherent: when (if) Marvell’s in-house photonics platform reaches commercial volume in 2027–2028+, Marvell could shift from “DSP-only supplier” to “DSP + integrated photonics” supplier — directly competitive with Coherent’s vertically integrated model.
The Marvell-Polariton trajectory is one of the multi-year displacement risks to Coherent’s vertical-integration thesis. It is not a 2026–2027 competitive threat, but it is a 2028+ structural watch item. ⚠ — directional inference; Marvell has not publicly disclosed a specific in-house photonics-product roadmap that competes with Coherent at the module level.
See LWLG IP / patents for the parallel electro-optic-polymer thesis at Lightwave Logic, which is in a similar long-cycle position vis-à-vis the InP incumbents.
Competitive moat summary
| Layer | Coherent moat strength | Erosion risk |
|---|---|---|
| Merchant InP EML | High — duopoly with Lumentum, scale + 6-inch wafer cost lead | Low through 2027; rising 2028+ as Lumentum Greensboro and Marvell-Polariton come online |
| VCSEL (Apple Face ID + datacom + automotive LiDAR) | High — Sherman TX fab, second-source position with Apple | Modest; Lumentum is original primary, Trumpf and others competing in automotive LiDAR |
| ROADM / WSS | High — duopoly with Lumentum, two-decade incumbency at telecom OEMs | Low; long product cycles |
| Pluggable transceiver modules | Moderate — vertically integrated but margin-pressured by Innolight/Eoptolink | High; persistent margin compression |
| SiC substrate | Moderate — long incumbency but commoditizing | High; Chinese capacity scaling, EV demand softening |
| Industrial fiber lasers | Moderate — broad product line but mature market | Moderate; Chinese competition |
The investment-grade summary: Coherent’s competitive position is strongest in InP EML / VCSEL source-lasers and ROADM/WSS, where the duopoly with Lumentum is structural and where the NVIDIA $2B investment specifically reinforces the franchise. The transceiver-module business is where Chinese competition exerts margin pressure. SiC and industrial lasers are mature, margin-pressured markets where Coherent’s competitive position is positive but the growth thesis is weak.
Caveats
- Module-share figures are ◐ — TrendForce / Cignal AI / IP-Fiber estimates, not company-disclosed. Treat as directional, not precise.
- The “Coherent vs Lumentum dual-supplier” framing is structurally correct but the relationship is genuinely dual-natured (mutual customer/supplier relationships exist alongside competition).
- Marvell-Polariton timing is highly uncertain — the forward in-house-photonics threat is real but the commercial-volume timing is 2028+ at the earliest.
- Apple is not a competitor — Apple is a customer (via Finisar Face ID second-source); see apple relationship.
Cross-links
- finisar acquisition — the 2019 deal that built the InP/VCSEL franchise
- coherent inc acquisition — the 2022 deal that brought the industrial-laser franchise
- inp eml process — Sherman TX 6-inch InP fab, the cost-lead specifics
- vcsel portfolio — VCSEL competitive set
- sic substrates — Wolfspeed/STMicro/TankeBlue/SICC competitive context
- nvidia partnership — the bilateral-duopoly preservation framing
- Lumentum competitors — the parallel duopoly counterpart
- Marvell ecosystem — Marvell as DSP partner today, Polariton as forward alternative
- LWLG IP / patents — EO-polymer parallel thesis
Sources
- COHR 10-K filings (Item 1 Business; competition discussion) — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000820318&type=10-K
- LITE 10-K filings — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001633978&type=10-K
- Wolfspeed (WOLF) 10-K filings — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000895419&type=10-K
- STMicroelectronics annual reports — https://investors.st.com/
- IPG Photonics 10-K filings — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001111928&type=10-K
- Innolight (300308.SZ) annual reports — http://www.innolight.com/
- Eoptolink (300502.SZ) annual reports — http://www.eoptolink.com/
- TrendForce / Cignal AI / IP-Fiber NVIDIA-bound 800G module supplier-share analysis (◐ industry attribution)
- Marvell-Polariton acquisition disclosures — https://www.marvell.com/
- Dell’Oro Group market reports (SiC, optical components) — https://www.delloro.com/