Photonic Computing Market (2026-2036)
The global Photonic Computing Market was valued at USD 0.42 billion in 2025. This market is expected to reach USD 14.56 billion by 2036 from an estimated USD 0.78 billion in 2026, registering a CAGR of 34.0% during the forecast period (2026-2036).
- Published
- Sep 2026
- Pages
- 314
- Format
- PDF + Excel
- Report ID
- MR-2233
- Base year
- 2025
- 2025 · BASELINE
- $420.0M
- 2036
- $14.56B
- CAGR 2026–2036
- 34.0%
2025 baseline · 2026–2036 forecast at 34.0% CAGR · hover a bar for the value
Key highlights
The global Photonic Computing Market is projected to reach USD 14.56 billion by 2036, driven by the AI power and bandwidth wall, semiconductor industry investment, and government funding.
North America is expected to account for the largest market share in 2026, while Asia-Pacific is projected to register the fastest growth during the forecast period.
Major semiconductor companies are acquiring photonic computing technology. In December 2025, Marvell agreed to acquire Celestial AI, developer of the Photonic Fabric optical interconnect platform, for upfront consideration of about USD 3.25 billion, with up to a further USD 2.25 billion payable if cumulative revenue milestones of USD 500 million to USD 2 billion are reached by the end of Marvell's fiscal 2029.
By offering, Optical Compute Interconnect is expected to account for the largest market share, whereas Photonic Processors & Accelerators are projected to witness the fastest growth through 2036.
AI's energy demand is pushing the industry toward light. The International Energy Agency projects data center electricity consumption to more than double from about 415 terawatt-hours in 2024 to around 945 terawatt-hours by 2030, and Marvell states that Celestial AI's optical interconnect delivers more than two times the power efficiency of copper interconnects.
Component supply is tight. Lumentum, a leading supplier of indium phosphide lasers used in optical interconnects, reported in 2026 a supply-demand imbalance of more than 30%, with its laser chips effectively sold out and its Japanese wafer fab fully allocated.
Report summary
| Particulars | Details |
|---|---|
| Forecast Period | 2026-2036 |
| Base Year | 2025 |
| Estimated Year | 2026 |
| CAGR (Value) | 34.0% |
| Format | PDF, Excel & Cloud Portal · 314 pages |
| Market Size (Value) in 2026 | USD 0.78 Billion |
| Market Size (Value) in 2036 | USD 14.56 Billion |
| Segments Covered | By Offering: Photonic Processors & Accelerators, Optical Compute Interconnect (Optical I/O Chiplets, Photonic Fabrics & Interposers, Co-Packaged Optics), Photonic Quantum Computers, Enabling Photonic Components. By Material Platform: Silicon Photonics, Indium Phosphide, Thin-Film Lithium Niobate, Silicon Nitride, Hybrid & Heterogeneous. By Application: AI Training & Inference, High-Performance Computing, Quantum Computing, Signal & Image Processing, Others. By Deployment: Hyperscale & Cloud Data Centers, HPC Centers & National Laboratories, Enterprise & Edge, Defense & Government. By End User: Cloud Service Providers, Semiconductor & System OEMs, Research Institutions & Government, Telecommunications Providers, Defense Organizations. |
| Countries Covered | North America: U.S., Canada. Europe: Germany, U.K., France, Netherlands, Belgium, Spain, Rest of Europe. Asia-Pacific: Taiwan, China, Japan, South Korea, Australia, Singapore, Rest of Asia-Pacific. Latin America: Brazil, Mexico, Rest of Latin America. Middle East & Africa: Israel, UAE, Saudi Arabia, Rest of Middle East & Africa. |
| Key Companies | NVIDIA, Broadcom, Marvell (Celestial AI), Intel, AMD, Cisco, TSMC, GlobalFoundries, Coherent, Lumentum, Lightmatter, Ayar Labs, Lightelligence, Q.ANT, Salience Labs, Optalysys, PsiQuantum, Xanadu, and Quandela. |
Report overview
Segments covered: offering, material platform, application, deployment, end user.
The growth of this market is mainly driven by the power and bandwidth limits of electronic interconnects in AI systems, large-scale investment and acquisitions by semiconductor companies, and government funding for photonics and quantum technologies. However, the precision and programmability limits of analog optical computing, the entrenched electronic computing ecosystem, constraints in laser and photonic component supply, and long commercialization timelines restrain the growth of this market.
Furthermore, optical scale-up interconnect for AI systems, photonic accelerators for energy-efficient inference, and photonic quantum computing are expected to offer growth opportunities for the stakeholders in this market. However, packaging and thermal integration of photonics with high-power processors, manufacturing yield and foundry capacity, standardization and interoperability, and the scale required for fault-tolerant photonic quantum computers remain major challenges impacting the growth of this market. Additionally, the commercialization of co-packaged optics, the emergence of optical I/O chiplets and memory disaggregation, consolidation of photonics startups into large semiconductor companies, and the adoption of new photonic materials are prominent trends in this market.
The Photonic Computing Market comprises technologies that use light, rather than or alongside electrical signals, to perform or directly enable computation. The market covers photonic processors and accelerators that execute mathematical operations such as matrix multiplication optically for AI and signal processing; optical compute interconnect that connects processors, memory, and switches within and between packages and racks, including optical I/O chiplets, photonic fabrics and interposers, and co-packaged optics integrated with switches and accelerators; photonic quantum computers that encode quantum information in photons; and enabling photonic components such as lasers, modulators, detectors, and photonic integrated circuits sold into these systems. Conventional pluggable optical transceivers for data center networking and telecommunications are excluded. The ecosystem spans photonic computing startups, semiconductor and networking companies, foundries and packaging providers, laser and component suppliers, hyperscalers, national laboratories, and governments.
The market is being pulled forward by the demands of AI. The International Energy Agency's April 2025 Energy and AI report estimated that data centers consumed about 415 terawatt-hours of electricity in 2024 and projected consumption to rise to around 945 terawatt-hours by 2030, and as AI systems link thousands of accelerators, copper interconnects face limits in bandwidth, reach, and power. In December 2025, Marvell agreed to acquire Celestial AI, whose Photonic Fabric technology delivers optical links directly into processors with more than twice the power efficiency of copper, for upfront consideration of about USD 3.25 billion, consisting of USD 1.0 billion in cash and about 27.2 million shares worth USD 2.25 billion, plus up to USD 2.25 billion more on revenue milestones. Marvell expects the business to begin contributing revenue in the second half of its fiscal 2028.
Photonic computing encompasses several technologies at different stages of maturity. Co-packaged optics and optical I/O are the nearest to volume deployment, with NVIDIA announcing silicon photonics-based networking switches with co-packaged optics in March 2025 and companies such as Ayar Labs developing optical I/O chiplets for processors. Photonic processors for AI have progressed from laboratory prototypes to systems demonstrated on real neural networks, including results published in Nature by Lightmatter in 2025, while companies such as Q.ANT are deploying photonic processing units in data center pilots. Photonic quantum computing developers, led by PsiQuantum, which raised USD 1 billion in 2025, are building toward fault-tolerant machines, supported by large government commitments.
The market faces significant constraints. Photonic devices depend on specialized lasers and materials, and indium phosphide laser capacity is tight: Lumentum reported in 2026 a supply-demand imbalance of more than 30%, with laser chips effectively sold out, while China added indium phosphide to its export control list in February 2025. Analog optical computing must overcome limits in precision and the need for electronic memory and conversion, and photonic computing must compete with a deeply entrenched electronic ecosystem, illustrated by NVIDIA's fiscal 2026 revenue of USD 215.9 billion. Commercialization timelines are long, as reflected in Marvell's expectation that Celestial AI's revenue will reach a USD 500 million annualized run rate only by the fourth quarter of fiscal 2028.
Market dynamics
18 factors across 5 forcesPower and Bandwidth Limits of Electronic Interconnects in AI Systems
The power and bandwidth limits of electronic interconnects in AI systems are a major factor driving the Photonic Computing Market. Training and serving large AI models requires connecting thousands of accelerators and large pools of memory with high bandwidth and low latency, but electrical links over copper lose signal integrity, consume more power, and shrink in reach as data rates rise. The International Energy Agency projects that data center electricity consumption will more than double from about 415 terawatt-hours in 2024 to around 945 terawatt-hours by 2030, driven by AI, making energy efficiency a central design goal. Marvell states that Celestial AI's Photonic Fabric delivers more than twice the power efficiency of copper interconnects, with longer reach and higher bandwidth, and that its 16-terabit-per-second chiplets can place optical connections directly into the processor rather than at the edge of the die. As AI clusters grow and copper approaches its limits, optical interconnect and photonic computing are expected to become essential elements of AI system design.
Large-Scale Investment and Acquisitions by Semiconductor Companies
Large-scale investment and acquisitions by semiconductor companies are significantly accelerating the commercialization of photonic computing. In December 2025, Marvell agreed to acquire Celestial AI for upfront consideration of about USD 3.25 billion, including USD 1.0 billion in cash, with up to a further USD 2.25 billion in stock payable if Celestial AI reaches cumulative revenue of at least USD 500 million, and in full above USD 2.0 billion, by the end of Marvell's fiscal 2029; Celestial AI had previously raised USD 250 million in March 2025. Lightmatter raised USD 400 million in 2024 at a valuation of USD 4.4 billion, and Ayar Labs raised USD 155 million in December 2024 with investment from NVIDIA, AMD, and Intel. With hyperscaler capital expenditure in 2026 approaching USD 700 billion, according to NVIDIA's chief financial officer, semiconductor and cloud companies are investing to secure optical technologies that can improve the performance and efficiency of their AI infrastructure.
Government Funding for Photonics and Quantum Technologies
Government funding for photonics and quantum technologies is supporting research, manufacturing, and early deployment. The Australian and Queensland governments announced a combined investment package of about AUD 940 million in 2024 to support PsiQuantum in building a utility-scale photonic quantum computer in Brisbane, and PsiQuantum is also developing a site in Chicago supported by the state of Illinois. The U.K. committed GBP 2.5 billion to its National Quantum Strategy in 2023, the European Union is funding integrated photonics pilot manufacturing lines under the Chips Act, and the U.S. supports integrated photonics manufacturing through the AIM Photonics institute and CHIPS Act programs. These investments reduce risk for private companies, build foundry and packaging infrastructure, and create early customers in national laboratories and defense.
Table of contents
13 chapters · 158 sections · 314 pages · click to expandSegmental analysis
| Segment | Largest share (2026) | Fastest growth (2026–2036) |
|---|---|---|
| By Offering | Optical Compute Interconnect | Photonic Processors & Accelerators |
| By Material Platform | Silicon Photonics | Thin-Film Lithium Niobate |
| By Application | AI Training & Inference | Quantum Computing |
| By Deployment | Hyperscale & Cloud Data Centers | Enterprise & Edge |
| By End User | Cloud Service Providers | Semiconductor & System OEMs |
By Offering
- The Optical Compute Interconnect segment is expected to account for the largest share of the market.
- The large share of this segment is mainly due to the early commercial deployment of co-packaged optics in AI networking and the urgency of overcoming copper interconnect limits.
- However, the Photonic Processors & Accelerators segment is projected to register the highest CAGR during the forecast period.
- The rapid growth of this segment is attributed to the transition of photonic processors from prototypes and pilots toward commercial inference and signal processing deployments.
By Material Platform
- The Silicon Photonics segment is expected to account for the largest share of the market.
- The large share of this segment is mainly due to its compatibility with CMOS manufacturing and its use in co-packaged optics and optical I/O.
- However, the Thin-Film Lithium Niobate segment is projected to register the highest CAGR during the forecast period.
- The rapid growth of this segment is attributed to its ultra-fast, low-power modulation capabilities for next-generation optical interconnect and computing.
By Application
- The AI Training & Inference segment is expected to account for the largest share of the market.
- The large share of this segment is mainly due to demand for optical interconnect in large AI clusters.
- However, the Quantum Computing segment is projected to register the highest CAGR during the forecast period.
- The rapid growth of this segment is attributed to large investments in photonic quantum computing and the expected deployment of utility-scale systems in the latter part of the forecast period.
By Deployment
- The Hyperscale & Cloud Data Centers segment is expected to account for the largest share of the market.
- The large share of this segment is mainly due to hyperscaler investment in optical interconnect for AI infrastructure, supported by capital expenditure approaching USD 700 billion in 2026.
- However, the Enterprise & Edge segment is projected to register the highest CAGR during the forecast period.
- The rapid growth of this segment is attributed to the potential of photonic processors for low-power inference and signal processing outside large data centers.
By End User
- The Cloud Service Providers segment is expected to account for the largest share of the market.
- The large share of this segment is mainly due to their role as the primary buyers of AI infrastructure.
- However, the Semiconductor & System OEMs segment is projected to register the highest CAGR during the forecast period.
- The rapid growth of this segment is attributed to the integration of optical chiplets and photonic fabrics into processors, switches, and systems.
Geographic analysis
North America
Largest shareIn 2026, North America is expected to account for the largest share of the global Photonic Computing Market. The region's dominance is supported by the leading AI infrastructure buyers, semiconductor companies, and photonic computing startups. The U.S. is home to NVIDIA, which announced co-packaged optics switches in March 2025, Marvell, which agreed to acquire Celestial AI for about USD 3.25 billion in December 2025, and Broadcom, as well as startups including Lightmatter, valued at USD 4.4 billion in 2024, Ayar Labs, which raised USD 155 million in December 2024, and PsiQuantum, which raised USD 1 billion in 2025 and is building a facility in Chicago. The AIM Photonics institute and CHIPS Act programs support U.S. photonics manufacturing, and hyperscalers' capital expenditure approaching USD 700 billion in 2026 creates the largest market for optical interconnect. Canada is home to Xanadu, a leading photonic quantum computing developer. North America
Europe
Europe is expected to account for a significant share of the market and has strong research and startup activity in photonic computing. Germany's Q.ANT is deploying photonic processing units in high-performance computing pilots, France's Quandela is developing photonic quantum computers, the U.K. is home to photonic computing companies such as Salience Labs and Optalysys and committed GBP 2.5 billion to its National Quantum Strategy in 2023, and Spain, the Netherlands, and Belgium have strong integrated photonics research and manufacturing ecosystems. The European Union is funding integrated photonics pilot lines under the Chips Act to strengthen European manufacturing capacity. The EU Chips Act, which aims to mobilize more than EUR 43 billion of public and private investment in semiconductors, includes support for photonics pilot lines and design capabilities, strengthening the European supply base for photonic computing. Europe
Asia-Pacific
Fastest growthHowever, Asia-Pacific is projected to register the highest CAGR during the forecast period. The rapid growth of this region is attributed to its central role in semiconductor manufacturing and advanced packaging, large AI infrastructure investment, and government-backed photonics and quantum programs. Taiwan's semiconductor and packaging industry is developing silicon photonics and co-packaged optics manufacturing, and Lumentum's laser fab in Japan is fully allocated to meet AI demand. Australia's federal and Queensland governments committed about AUD 940 million in 2024 to support PsiQuantum's photonic quantum computer in Brisbane, China has active photonic computing research at universities and companies, and Japan's NTT is advancing all-photonics network technology through its IOWN initiative. Asia-Pacific
Latin America
Latin America is expected to account for a smaller share of the market. Brazil has established photonics research centers and a growing data center sector, supported by a national AI plan launched in 2024 with planned investment of about BRL 23 billion, and Mexico's growing role in electronics and AI server assembly, including Foxconn production in Guadalajara, could create demand for photonic components over time. Adoption in the region is expected to follow deployments of optical interconnect in AI data centers built by global cloud providers. Latin America
Middle East & Africa
The Middle East & Africa is expected to register strong growth from a small base. Israel is a significant center of photonics and networking innovation, including NVIDIA's networking operations, which developed its co-packaged optics switches, and several optical interconnect startups. The UAE and Saudi Arabia are investing heavily in AI infrastructure, including the 1-gigawatt Stargate UAE cluster announced in May 2025 and Saudi Arabia's Humain, and in quantum research through institutions such as the UAE's Technology Innovation Institute, creating future demand for optical interconnect and photonic computing technologies. Stargate UAE's initial 1-gigawatt cluster is part of a planned 5-gigawatt AI campus in Abu Dhabi, and Saudi Arabia's Humain agreed in May 2025 to purchase 18,000 NVIDIA GB300 chips for its initial deployment, indicating the scale of AI infrastructure that will require advanced interconnect in the region. Middle East & Africa
Competitive landscape
The global Photonic Computing Market is at an early stage and includes large semiconductor and networking companies commercializing co-packaged optics and optical interconnect, specialized photonic computing and optical I/O startups, photonic quantum computing developers, laser and component suppliers, and foundries building photonics manufacturing capacity. Competition centers on energy efficiency, bandwidth and latency, integration with processors and switches, thermal reliability, manufacturability and cost, software compatibility, and access to hyperscaler customers.
Leading companies are acquiring or investing in photonics startups, developing co-packaged optics and optical chiplets, securing laser and packaging capacity, and partnering with hyperscalers and foundries. Marvell's acquisition of Celestial AI and strategic investments by NVIDIA, AMD, and Intel in Ayar Labs illustrate the convergence of electronic and photonic computing.
The report provides a comprehensive competitive assessment of the leading companies operating in the global Photonic Computing Market. The key players profiled in the report include NVIDIA Corporation (U.S.), Broadcom Inc. (U.S.), Marvell Technology, Inc. (Celestial AI) (U.S.), Intel Corporation (U.S.), Advanced Micro Devices, Inc. (U.S.), Cisco Systems, Inc. (U.S.), Taiwan Semiconductor Manufacturing Company Limited (Taiwan), GlobalFoundries Inc. (U.S.), Coherent Corp. (U.S.), Lumentum Holdings Inc. (U.S.), Lightmatter, Inc. (U.S.), Ayar Labs, Inc. (U.S.), Lightelligence (U.S./China), Q.ANT GmbH (Germany), Salience Labs Ltd. (U.K.), Optalysys Ltd. (U.K.), PsiQuantum Corp. (U.S.), Xanadu Quantum Technologies Inc. (Canada), and Quandela SAS (France).
- NVIDIA
- Broadcom
- Marvell (Celestial AI)
- Intel
- AMD
- Cisco
- TSMC
- GlobalFoundries
- Coherent
- Lumentum
- Lightmatter
- Ayar Labs
- Lightelligence
- Q.ANT
- Salience Labs
- Optalysys
- PsiQuantum
- Xanadu
- Quandela
Expert perspectives
Photonic computing is moving from research to commercial reality, led not by optical processors replacing GPUs but by light replacing copper inside AI systems. Marvell's agreement to pay about USD 3.25 billion upfront for Celestial AI, NVIDIA's co-packaged optics switches, and the investments of NVIDIA, AMD, and Intel in Ayar Labs show that the industry's largest companies see optical interconnect as essential to scaling AI within power limits that the International Energy Agency projects will more than double data center electricity demand by 2030.
Three structural changes are expected to shape the market through 2036. First, optical interconnect will migrate from switches to processors, through co-packaged optics, optical I/O chiplets, and photonic fabrics, becoming a standard part of AI system design. Second, photonic processors will find commercial niches in energy-efficient inference and signal processing, working alongside electronic accelerators rather than replacing them. Third, photonic quantum computing will progress toward utility-scale machines, supported by government funding, creating a large but longer-term market.
For companies planning entry or expansion, the most attractive positions over the forecast period are likely to be found in optical scale-up interconnect, co-packaged optics and optical I/O chiplets, laser and photonic component supply, photonic packaging and test, and photonic accelerators for inference. The principal risks are the precision limits of analog optical computing, the entrenched electronic ecosystem, laser supply constraints, and long commercialization timelines.
Customer perspectives
Insights gathered during primary interviews with hyperscaler infrastructure architects, semiconductor executives, HPC center directors, and photonics engineers highlight where purchasing priorities are shifting. The following perspectives reflect recurring themes raised across these discussions.
“This reflects near-term demand for optical scale-up interconnect and the importance of reliability and supply diversity.”
“This indicates the complementary role of photonic accelerators and the importance of software.”
“This points to laser supply as a gating factor for photonic computing adoption.”
Frequently asked questions
The global Photonic Computing Market is estimated at USD 0.78 billion in 2026.
Cite this report
Meticulous Research. (2026). Photonic Computing Market - Opportunity Analysis and Industry Forecast (2026-2036) (Report No. MR-2233). Meticulous Market Research Pvt. Ltd. https://www.meticulousresearch.com/reports/photonic-computing-market-6916