What are the fundraising trends in the semiconductor industry?
In our semiconductor industry deck, you will find everything you need to understand the market
SUMMARY
We analyzed publicly disclosed equity and equity-like growth rounds raised by pure-play semiconductor device and wafer-fab or foundry companies between January 2026 and July 2026, using a minimum disclosed deal size of $300K and excluding equipment, materials, EDA, IP-only, packaging/test, distribution, EMS, end-product OEMs, grants, debt, subsidies, and undisclosed-size rounds. The resulting YTD 2026 sample contains 32 deals across 32 unique companies, with $9.0B of disclosed capital raised.
The semiconductor market is taking in far more capital than in the comparable YTD windows of 2025 and 2024, but the expansion is not coming from more deals. YTD 2026 capital reached $9.0B, compared with $1.3B in YTD 2025 and $1.8B in YTD 2024, while deal count fell to 32 from 48 in YTD 2025.
The main story is larger rounds, not broader fundraising. The median semiconductor round rose to $60.0M in YTD 2026, up from $18.5M in YTD 2025 and $23.3M in YTD 2024, while the average round reached $281.9M because several billion-dollar and half-billion-dollar financings pulled the market upward.
Capital in the semiconductor market is highly concentrated. The top 10 YTD 2026 rounds captured 86.5% of all dollars, and the bottom half of deals captured only 4.1%, which means the headline funding total mostly describes a small set of strategically important companies.
Logic Semiconductors dominate deal activity, with 23 of 32 YTD 2026 deals, but not the full capital story. Memory Semiconductors and Foundry Services produced only 3 deals combined yet absorbed 48.0% of all capital, showing that investors are assigning premium value to manufacturing capacity, memory supply, and infrastructure bottlenecks.
Late-stage and growth rounds control the semiconductor market by dollars. Series B and later rounds, including Growth Equity, captured 88.5% of YTD 2026 capital, while Seed, Series A, and Unknown stages together captured only 11.5%.
At the same time, company formation has not disappeared. Series A was the most common stage by deal count, with 14 deals, and Seed added 3 more, but the capital behind these younger companies remains much smaller than the capital flowing to proven platforms, national champions, and supply-chain-critical assets.
Asia-Pacific led the semiconductor market by capital in YTD 2026, with 54.3% of dollars, driven by Nanya Technology, Rapidus, Rebellions, and multiple Indian and Korean fabless rounds. North America remained close behind on deal count and produced the major AI accelerator and photonics financings, while Europe showed fewer but unusually large rounds.
First financings were relatively scarce. Only 5 of 32 YTD 2026 deals were first financings, and those rounds captured just 6.2% of capital, which confirms that semiconductor investors are mostly compounding capital into companies that already have technical proof, manufacturing access, customer pull, or strategic relevance.
The practical interpretation is clear: the semiconductor market is hot, but not evenly hot. It is strongest where the company can point to a hard infrastructure bottleneck, such as AI inference, DRAM capacity, leading-edge foundry access, co-packaged optics, optical interconnects, or power delivery for AI data centers.
This chart, featured in our semiconductor industry deck, shows the revenue mix across customer segments in the semiconductor industry
Is more or less capital going into the semiconductor market?
More capital is going into the semiconductor market in YTD 2026, and the increase is dramatic. The market raised $9.0B across 32 deals between January and July 2026, compared with $1.3B across 48 deals in the comparable 2025 window and $1.8B across 34 deals in the comparable 2024 window.
The direct answer is that semiconductor funding is up by dollars but down by deal count versus YTD 2025. That matters because it means the market is not simply becoming more active across the board; it is becoming more capital-intensive at the top.
The largest YTD 2026 rounds explain the change. Nanya Technology raised $2.5B, Rapidus raised $1.7B, and Cerebras Systems raised $1.0B. Together, the top 3 rounds represented 57.7% of all YTD 2026 capital.
Even after removing the largest deal, the semiconductor market still raised $6.5B in YTD 2026, far above the full YTD 2025 total. That confirms the increase is not just one outlier; it is a broader shift toward very large strategic financings.
The real signal is that semiconductor capital has moved from venture-style company funding into infrastructure-scale funding. AI compute, memory capacity, optical I/O, and leading-edge foundry access are being financed more like strategic supply-chain assets than ordinary startups.
Is semiconductor funding driven by more deals or larger rounds?
Semiconductor funding in YTD 2026 is being driven by larger rounds, not more deals. Deal count fell from 48 in YTD 2025 to 32 in YTD 2026, but capital rose from $1.3B to $9.0B.
The median round size tells the cleanest story. The median semiconductor round increased to $60.0M in YTD 2026, up from $18.5M in YTD 2025 and $23.3M in YTD 2024, so the typical qualifying round became much larger.
The average round moved even more sharply, reaching $281.9M in YTD 2026 versus $27.9M in YTD 2025. That average is not representative of a normal semiconductor startup, but it is useful because it shows how much the top of the market has changed.
The $50M threshold is the structural dividing line. In YTD 2026, 18 of 32 deals exceeded $50M, and 14 deals exceeded $100M. In YTD 2025, only 5 deals exceeded $50M and only 1 exceeded $100M.
For more detail on how round sizes are reshaping the semiconductor market, see the full semiconductor market report.
Is semiconductor capital moving toward later-stage or earlier-stage companies?
Semiconductor capital is moving strongly toward later-stage and growth companies by dollars, even though Series A remains active by deal count. In YTD 2026, late-stage rounds, defined as Series B and later including Growth Equity, captured 88.5% of all capital.
Growth Equity alone represented $5.35B, or 59.3% of capital, across 5 deals. Series D+ added another $1.5B, while Series B contributed $1.1B. That means the dollar center of gravity sits well beyond company formation.
But the deal-count picture is more balanced. Series A was the largest stage by frequency, with 14 of 32 deals, while Seed added 3 deals and Unknown added 2. The semiconductor market is still forming companies, but those companies are not where most dollars are going.
This creates a barbell structure. At one end, there is broad Series A formation across AI chips, photonics, sensors, power chips, and regional fabless startups. At the other end, billion-dollar and half-billion-dollar rounds are going to companies with manufacturing paths, customer pull, or national strategic relevance.
The practical takeaway is that early-stage semiconductor startups can still raise, but the market’s biggest checks are reserved for companies that have already crossed a proof threshold. In semiconductors, that proof is usually silicon feasibility, a credible manufacturing path, strong customer demand, or strategic supply-chain importance.
This chart, featured in our semiconductor industry deck, compares the main business model options for fabless semiconductor companies
Is the semiconductor market maturing or still experimental?
The semiconductor market is maturing by capital allocation, but it remains experimental by company formation. The most important YTD 2026 dollars went to large, strategic, follow-on rounds, while most of the deal count still came from younger and narrower technology bets.
The maturity signal is clearest in the late-stage capital split. Series B and later rounds captured 88.5% of YTD 2026 dollars, and the top categories by capital included Logic Semiconductors, Memory Semiconductors, and Foundry Services rather than only small exploratory niches.
Foundry Services and Memory Semiconductors are especially mature signals. Foundry Services produced just 1 deal but captured 18.8% of capital through Rapidus, while Memory Semiconductors produced only 2 deals but captured 29.2% of capital through Nanya Technology and XCENA.
The experimental layer is visible in the long tail. The bottom half of deals captured only 4.1% of capital, and many sub-$20M rounds were productization bets in areas like Indian fabless chips, analog-in-memory compute, edge AI SoCs, wireless chips, and sensor semiconductors.
The honest interpretation is that the semiconductor market is mature where it touches national infrastructure and AI supply bottlenecks, but still experimental where companies are trying to prove new architectures or narrow device categories. Both are happening at the same time.
Are new startups still entering the semiconductor market?
Yes, new startups are still entering the semiconductor market, but fewer first financings are appearing in YTD 2026 than in the comparable 2025 and full-year 2025 periods. Only 5 of 32 YTD 2026 deals were first financings, equal to 15.6% of deal count.
That is lower than YTD 2025, when first financings represented 25.0% of deals, and lower than full-year 2025, when they represented 28.0%. So company formation has not stopped, but the market is less seed-heavy than it was last year.
The first financings that did appear were meaningful. Kandou AI raised $225M, OLIX raised $220M, Reed Semiconductor raised $100M, Chiral raised $12M, and Mosaic SoC raised $3.8M. This is not a market where every first financing is tiny.
Still, first financings captured only 6.2% of YTD 2026 capital. That means new entrants are visible in the semiconductor market, but the overwhelming majority of dollars are still going to follow-on companies with prior proof.
For a broader view of first financings, follow-on rounds, and company formation in the semiconductor market, see the semiconductor market deck.
Are more investors entering the semiconductor market?
More investors are participating in the semiconductor market than the deal count alone might suggest, but the 2026 investor base is highly concentrated around strategic and institutional backers. The YTD 2026 dataset includes approximately 98 disclosed named investors and about 54 unique tier-1 or strategic investors.
That is fewer disclosed investors than full-year 2025, which had approximately 253 named investors, but the comparison is not apples-to-apples because 2026 is still incomplete. More importantly, several 2026 rounds did not fully disclose their syndicates, so repeat investor counts are conservative.
The names that matter most are not only venture funds. Strategic investors such as NVIDIA, AMD, Cisco Systems, SanDisk, Kioxia, Solidigm, SoftBank, Synopsys, Cadence, Alchip, MediaTek, and Samsung Catalyst Fund show up across major rounds.
That strategic participation is a major difference between semiconductors and many software markets. In the semiconductor market, a credible strategic investor can validate manufacturing relevance, ecosystem fit, customer pull, or supply-chain importance more strongly than a financial investor alone.
The practical takeaway is that investor breadth is present, but logo quality matters more than raw investor count. A semiconductor round backed by customers, suppliers, hyperscalers, or major chip ecosystem players often carries a stronger signal than a round backed only by generalist capital.
This chart, featured in our semiconductor industry deck, shows annual funding in semiconductor startups
Are top investors getting more or less active in semiconductor?
Top investors remain active in semiconductor, but the pattern is more strategic and company-specific than broadly thematic. In YTD 2026, only a small set of investors appeared in more than one disclosed round, including NVIDIA, AMD, Matter Venture Partners, European Innovation Council Fund, Ireland-linked state capital, and Vertex Ventures.
Compared with full-year 2025, the repeat-investor list is shorter because the current period is only partial and because many large 2026 rounds did not fully disclose all participants. In 2025, repeat names included Maverick Silicon, Atreides Management, Tiger Global, Valor Equity Partners, EIC Fund, Bosch Ventures, Mayfield, Sutter Hill Ventures, Koch Disruptive Technologies, and several others.
The more important point is that the top investors are not behaving like a single homogeneous semiconductor VC cluster. NVIDIA and AMD are validating AI infrastructure and photonics; EIC Fund and Ireland-linked state capital are supporting European deep-tech capacity; Matter Venture Partners appears around photonics and chiplet-related semiconductor assets.
Strategic investor activity is therefore more informative than the number of repeat financial investors. In semiconductors, the best signal is often whether the round includes a strategic party that has a reason to care about deployment, supply, manufacturing, or ecosystem integration.
The honest interpretation is that top investors are still active, but they are selective. They are not funding every chip narrative; they are backing companies that map to concrete bottlenecks in AI compute, interconnect, memory, power, or sovereign manufacturing.
Which semiconductor subcategories are gaining momentum?
Memory Semiconductors, Foundry Services, and selected Logic Semiconductor segments are the subcategories gaining the most momentum by capital in YTD 2026. Memory and foundry together produced only 3 deals but captured 48.0% of all dollars, which is the clearest sign that capital is moving toward strategic infrastructure bottlenecks.
Memory Semiconductors rose from 1.5% of YTD 2025 capital to 29.2% of YTD 2026 capital. Nanya Technology’s $2.5B private placement and XCENA’s $135M Series B show how AI demand is repricing both conventional memory capacity and memory-centric computing.
Foundry Services also shifted sharply. The category had only $14.9M in YTD 2025 and no YTD 2024 capital, but Rapidus raised $1.7B in YTD 2026 for leading-edge 2nm foundry capacity. That is not ordinary venture momentum; it is national industrial strategy showing up in a private funding dataset.
Within Logic Semiconductors, the strongest gaining pockets are AI inference accelerators, photonic chips, optical I/O, and interconnect. Cerebras, MatX, Etched, Ayar Labs, Rebellions, Axelera AI, Positron AI, OLIX, Fractile, Neurophos, OpenLight, and Xscape Photonics all point to the same practical issue: data movement and AI compute efficiency have become investable constraints.
We cover these subcategory shifts in more detail in the deeper analysis of the semiconductor market.
Which semiconductor subcategories are losing momentum?
Sensor Semiconductors, Analog Semiconductors, and Power Semiconductors are losing momentum by capital share in YTD 2026, even though some of those areas remain strategically important. The decline is relative, not absolute: the market is being overwhelmed by very large logic, memory, and foundry rounds.
Sensor Semiconductors fell from 9.2% of YTD 2025 capital to 0.2% in YTD 2026. The category still had 2 deals, but they were small compared with AI accelerator, memory, and foundry financings.
Analog Semiconductors fell from 5.6% of YTD 2025 capital to 0.1% in YTD 2026, with only one $10M qualifying round. That does not mean analog chips are unimportant; it means the disclosed equity market is currently pricing other bottlenecks more aggressively.
Power Semiconductors are the more interesting case. Power chips produced 3 deals in YTD 2026, including Reed Semiconductor at $100M, but the category still captured only 1.6% of capital. Given the power-delivery needs of AI data centers, that gap suggests investors recognize the problem but have not yet priced most power-chip startups like platform-level infrastructure companies.
The practical interpretation is that categories are not losing technical relevance; they are losing competition for capital attention. In 2026, the dollars are chasing AI compute, memory bandwidth, optical movement of data, and leading-edge capacity first.
This chart, featured in our semiconductor industry deck, looks at TSMC’s strategy in semiconductors
Which regions are gaining momentum in semiconductor funding?
Asia-Pacific is the region gaining the most momentum in semiconductor funding by capital. It captured 54.3% of YTD 2026 dollars, up from only 3.7% in YTD 2025 and 16.7% in YTD 2024.
The reason is not simply more APAC startups. Asia-Pacific had 13 of 32 YTD 2026 deals, but the capital surge came mainly from large strategic rounds: Nanya Technology in Taiwan, Rapidus in Japan, Rebellions and XCENA in South Korea, and a long tail of Indian fabless companies.
Europe is also gaining momentum by median round quality. It produced 7 YTD 2026 deals and $990.8M of capital, with an unusually high median regional deal size of $220M, driven by Axelera AI, Kandou AI, OLIX, Fractile, Equal1, Chiral, and Mosaic SoC.
North America remains strong rather than newly gaining. It captured 34.7% of YTD 2026 capital and 37.5% of deals, with major AI and photonics rounds from Cerebras Systems, MatX, Etched, Ayar Labs, Positron AI, Neurophos, Reed Semiconductor, OpenLight, Xscape Photonics, Amber Semiconductor, SPARK Microsystems, and Efficient Computer.
For regional tracking across Asia-Pacific, North America, Europe, and emerging semiconductor geographies, see the market report covering semiconductor funding by region.
Which regions are losing momentum in semiconductor funding?
North America is losing capital share versus YTD 2025, even though it remains one of the strongest semiconductor regions. Its share fell from 72.4% of YTD 2025 capital to 34.7% of YTD 2026 capital, mostly because Asia-Pacific produced several billion-dollar and large strategic rounds.
This is not a North American collapse. North America still raised $3.1B in YTD 2026, which is more than three times its YTD 2025 total of $970.7M. The region lost share because the whole market grew around APAC infrastructure and capacity financings.
The Middle East is the clearest region losing momentum under the strict company-origin definition. It had 9.3% of YTD 2024 capital and 3.3% of YTD 2025 capital, but zero qualifying YTD 2026 deals.
Latin America and Africa also had no qualifying deals in YTD 2026, matching the absence in the earlier datasets. That absence is informative because semiconductor device and foundry fundraising still clusters where there are dense engineering labor pools, supply-chain adjacency, national industrial policy, or hyperscale AI demand.
The practical takeaway is that regional share changes should be read carefully. North America remains strong in absolute dollars, Europe is showing fewer but larger bets, and Asia-Pacific is where sovereign-capacity and memory/foundry financing changed the shape of the market.
Is semiconductor becoming more global or regionally concentrated?
The semiconductor market is becoming more global by deal geography, but still regionally concentrated in the places that can support chip design, manufacturing, customers, and industrial policy. YTD 2026 qualifying deals came from Asia-Pacific, North America, and Europe, with no deals in Latin America, the Middle East, or Africa.
Asia-Pacific led YTD 2026 capital with 54.3%, North America followed with 34.7%, and Europe captured 11.0%. That looks more balanced than a North America-only market, but it is still concentrated across three semiconductor ecosystems.
The deal-count split also shows concentration. Asia-Pacific had 13 deals, North America had 12, and Europe had 7. No other region contributed a qualifying semiconductor-device or foundry round under the strict definition.
The global story is therefore nuanced. More countries are producing semiconductor companies, especially India, Korea, Japan, Taiwan, the United States, Canada, the United Kingdom, Ireland, the Netherlands, and Switzerland, but the capital pools still follow places with deep technical talent, fabs or fabless ecosystems, and strategic customers.
The real signal is that semiconductor globalization is not evenly distributed. It is a three-region market with increasingly different roles: Asia-Pacific for memory, foundry, and fabless formation; North America for AI accelerator and photonics scale; Europe for fewer but credible deep-tech infrastructure bets.
This chart, featured in our semiconductor industry deck, shows how advanced node demand has driven growth in the semiconductor industry over time
Is semiconductor capital moving toward proven winners or new opportunities?
Semiconductor capital is moving toward proven winners and strategically validated companies, not primarily toward new opportunities. Follow-on rounds represented 84.4% of YTD 2026 deals and captured 93.8% of capital.
First financings were only 15.6% of YTD 2026 deals and just 6.2% of capital. That is a strong signal that investors are not broadly seeding brand-new semiconductor companies at scale.
The companies receiving the largest checks had clear proof points. Nanya is an established DRAM manufacturer, Rapidus is a national-champion foundry project, Cerebras has wafer-scale AI processors, Ayar Labs is moving toward volume production of co-packaged optics, and Etched is tied to transformer-specific AI inference demand.
This does not mean new opportunities are absent. Kandou AI and OLIX both counted as first financings and raised more than $200M each, which shows that a new semiconductor company can raise a very large first round if it maps to a critical bottleneck.
The best reading is that the semiconductor market is opportunity-seeking but proof-hungry. New concepts can still raise, but they need to look less like a slide-deck idea and more like a supply-chain, manufacturing, compute, or interconnect answer.
Our full market view on semiconductor follow-on funding tracks which companies are converting technical relevance into repeat capital.
Is the semiconductor market becoming winner-takes-most?
The semiconductor market is not strictly winner-takes-all, but it is clearly winner-takes-most by capital. In YTD 2026, the top 10 rounds captured 86.5% of all dollars, while the bottom half of deals captured only 4.1%.
The concentration is stronger than in YTD 2025, when the top 10 rounds captured 57.9% and the bottom half captured 14.0%. It is also stronger than full-year 2025, when the top 10 captured 59.3% and the bottom half captured 7.7%.
The top 1 deal alone, Nanya Technology, captured 27.7% of YTD 2026 capital. The top 3 deals, Nanya, Rapidus, and Cerebras Systems, captured 57.7%. That is a high concentration level, but it reflects infrastructure scarcity rather than simple hype.
The largest-to-median ratio makes the point more sharply. The largest YTD 2026 deal was 41.7 times the median round, compared with 13.5 times in YTD 2025. The typical company is not experiencing the same market as the top strategic assets.
The practical takeaway is that aggregate semiconductor funding totals are useful only when paired with concentration metrics. A rising total can coexist with a difficult market for smaller chip startups if most dollars are flowing into a small number of capacity, memory, AI, or foundry winners.
Is the next wave of semiconductor winners becoming visible?
The next wave of semiconductor winners is becoming visible, but it is easier to see by bottleneck than by company stage. The strongest candidates are clustered around AI inference, photonics, optical interconnect, memory-centric computing, power delivery, and sovereign foundry capacity.
Several YTD 2026 names already look like visible next-wave contenders because they raised large rounds in strategically important areas. MatX and Positron AI point to AI inference competition; Etched points to transformer-specific ASICs; Ayar Labs, Xscape Photonics, OpenLight, Neurophos, OLIX, and Fractile point to optical and photonic constraints.
Memory and foundry winners are also visible, but in a different way. Nanya and Rapidus are not typical startups; they are capacity and industrial-policy assets. Their presence shows that the next wave of semiconductor winners may include national champions and manufacturing platforms, not just venture-backed fabless designers.
India’s fabless pipeline is another early signal. C2i Semiconductors, HrdWyr, Vervesemi, BigEndian, Aheesa, and optoML raised mostly sub-$20M rounds, which suggests active formation but not yet global-scale capital access.
The honest interpretation is that the next wave is visible at the theme level, but not fully sorted at the company level. The winners will likely be the teams that can turn a narrow technical bottleneck into shipped silicon, production partnerships, and repeat customer demand.
As this chart shows, and as featured in our semiconductor industry deck, search interest in semiconductors has been rising steadily
Is the semiconductor funding landscape fragmenting or consolidating?
The semiconductor funding landscape is fragmenting by company type but consolidating by capital allocation. The YTD 2026 dataset spans AI accelerators, photonics, memory, power, sensors, analog, quantum processors on CMOS, foundry services, wireless chips, and edge SoCs, but the money is concentrated in a much smaller set of strategic themes.
Logic Semiconductors alone produced 23 of 32 deals, which shows wide experimentation inside the category. But Logic’s capital-share-to-deal-share ratio was only 0.70, meaning it had many deals but did not absorb capital in proportion to its frequency.
Foundry Services and Memory Semiconductors show the opposite pattern. Their capital-share-to-deal-share ratios were 6.03 and 4.67, respectively. That is consolidation around the scarcest infrastructure assets.
The market is also consolidating around the $50M threshold. Eighteen YTD 2026 deals exceeded $50M and accounted for 97.2% of all capital. Below that line, the rounds are mostly technology-discovery and productization bets; above it, they are manufacturing, deployment, or strategic-supply-chain bets.
The result is a two-speed semiconductor market. Many companies are exploring specific architectures, but the capital market is consolidating around companies that look indispensable to AI infrastructure, memory supply, optical data movement, power delivery, or advanced manufacturing.
Where is investor attention shifting in semiconductor?
Investor attention in the semiconductor market is shifting toward concrete bottlenecks in AI infrastructure, not generic chip branding. The strongest YTD 2026 rounds named specific constraints: transformer inference, wafer-scale compute, DRAM capacity, 2nm foundry access, co-packaged optics, photonic AI acceleration, memory-centric computing, and power delivery.
That is why the market rewarded companies such as Cerebras Systems, MatX, Etched, Ayar Labs, Rebellions, Nanya Technology, Rapidus, Axelera AI, Positron AI, OLIX, Fractile, and Neurophos. These companies are not just “AI chip” stories; they are attempts to solve compute, bandwidth, energy, or supply constraints that customers can understand.
Strategic participation confirms the shift. AMD and NVIDIA appeared in Ayar Labs, Cisco and memory players appeared in Nanya, SoftBank, Synopsys, Cadence, and Alchip appeared in Kandou AI, and NVIDIA appeared in Xscape Photonics. The practical signal is that investors care deeply about ecosystem validation.
Power delivery is also emerging, though it has not yet received the same capital intensity. Reed Semiconductor, Amber Semiconductor, and C2i Semiconductors show that investors recognize power as an AI data-center bottleneck, but the category still captured only 1.6% of YTD 2026 capital.
For continued tracking of where investor attention is moving across AI accelerators, memory, foundry, photonics, power, analog, and sensor chips, see the semiconductor market report.
INSIGHTS
The insights below come from reviewing every disclosed YTD 2026 equity round in the semiconductor market and comparing the pattern with 2024 and 2025 semiconductor funding data.
- The semiconductor market looks broad by company count but narrow by capital allocation. Thirty-two companies raised in YTD 2026, yet the top 10 rounds captured 86.5% of all dollars. That means the phrase “semiconductor funding is hot” is mostly true for companies that already look strategically indispensable.
- The median round was $60.0M while the average was $281.9M, so the average is not a realistic benchmark for most founders. A company building a small fabless product should not compare itself with a sovereign foundry project, a DRAM capacity raise, or a wafer-scale AI processor company.
- Memory and foundry produced only 3 of 32 deals but absorbed 48.0% of all capital. The market is not rewarding deal frequency; it is rewarding infrastructure bottlenecks that cannot be solved with software or small design teams.
- Logic Semiconductors dominated deal count with 23 deals, but its 50.0% capital share was below its 71.9% deal share. This suggests Logic is the broad experimentation layer, while memory and foundry are the strategic-capex layer.
- Foundry Services had the highest capital-share-to-deal-share ratio in the dataset, at 6.03. One Rapidus round outweighed most individual deals, which shows that leading-edge wafer access is being valued as national infrastructure rather than as a normal venture segment.
- Memory Semiconductors had a capital-share-to-deal-share ratio of 4.67. That is a clear sign that AI demand is repricing memory capacity, even though there are very few pure-play memory companies available for investors to back.
- The bottom half of deals accounted for only 4.1% of YTD 2026 dollars. The long tail matters for technology discovery, but it barely moves aggregate funding statistics. Any market-size claim based only on total dollars will mostly describe the largest industrial-policy and AI-infrastructure bets.
- Early-stage companies still matter, but they are not where the dollars are concentrated. Seed, Series A, and Unknown rounds represented a meaningful share of deal count, yet only 11.5% of YTD 2026 capital. In semiconductor, new company formation and capital intensity are separate stories.
- First financings were only 15.6% of deals and 6.2% of capital. That suggests capital is compounding into known teams rather than broadly seeding fresh semiconductor entrants. A prior proof point is becoming a funding gate, not just a valuation enhancer.
- Asia-Pacific’s YTD 2026 capital lead is not just about startup volume. The region has national champions, memory assets, foundry projects, and Korean AI-chip scale-ups that can justify very large rounds. That mix is structurally different from a market made only of small fabless startups.
- North America had a higher median deal size than Asia-Pacific despite lower total capital. Asia-Pacific’s total was pulled up by Nanya and Rapidus, while its median was pulled down by smaller Indian and Korean fabless rounds. North America’s sample was more consistently heavy with $100M-plus AI and photonics rounds.
- Europe’s $220M median deal size is unusually high because its sample is concentrated in OLIX, Fractile, Axelera AI, Kandou AI, and Equal1. The European signal is therefore “few but credible infrastructure-scale bets,” not a deep wave of small chip-company formation.
- Power Semiconductors are underpriced relative to the importance of the bottleneck. AI data centers clearly need better power delivery, yet Power Semiconductors captured only 1.6% of YTD 2026 capital. The gap suggests room for stronger funding if power-chip companies can prove platform-level relevance.
- Photonics appears repeatedly across Ayar Labs, Xscape Photonics, OpenLight, Neurophos, OLIX, Fractile, and related optical-compute or connectivity companies. Investors are treating data movement and interconnect energy as bottlenecks almost as important as compute itself.
- Strategic investors are unusually important in the strongest rounds. AMD, NVIDIA, Cisco, memory players, SoftBank, Synopsys, Cadence, Alchip, Samsung, and MediaTek all appear in the ecosystem. In this market, strategic participation is often a stronger validation signal than a high nominal valuation alone.
- The strongest semiconductor rounds usually combine three proofs: silicon feasibility, manufacturing path, and customer pull. A concept-only architecture is much harder to fund at scale than a company with working silicon, credible production partners, and a clearly named bottleneck.
- Indian fabless companies appear frequently in the sub-$20M band. That suggests India is moving from semiconductor-policy rhetoric into product-company formation, but the capital scale remains far below the United States, Europe, Korea, Taiwan, and Japan.
- The stage distribution is a barbell. Series A led deal count, while Growth Equity and Series D+ dominated capital. The semiconductor market is simultaneously early in breadth and late-stage in capital intensity.
- No qualified deals appeared in Latin America, the Middle East, or Africa under the strict pure-play definition in YTD 2026. The absence is meaningful: semiconductor device and foundry fundraising remains concentrated where there are deep engineering labor pools, supply-chain adjacency, national industrial policy, or hyperscale AI customers.
- The dataset rewards vertical specificity more than generic “AI chip” branding. Companies that identified a concrete bottleneck, such as transformer inference, co-packaged optics, DRAM capacity, voltage regulation, low-power wireless, quantum processors on CMOS, or 2nm foundry access, raised more credible rounds than generic chip-design narratives.
This chart, featured in our semiconductor industry deck, shows how advanced foundry node manufacturing technology has evolved over time
OUR METHODOLOGY TO BUILD THIS TRACKER
We built this semiconductor funding tracker by reviewing publicly disclosed equity and equity-like growth rounds raised by pure-play semiconductor device and wafer-fab or foundry companies between January 2026 and July 2026. A company counts as pure-play when more than 80% of its activity is dedicated to selling semiconductor devices or semiconductor wafer manufacturing services.
We applied four core filters to build the dataset. First, we only included equity rounds, private placements, and growth equity rounds, while excluding grants, debt, subsidies, structured financings, acquisitions, SPAC transactions, business combinations, and IPOs unless the raw data explicitly treated them as qualifying equity rounds. Second, we only counted disclosed rounds of $300K or more. Third, we only kept pure-play semiconductor companies. Fourth, every entry had to be confirmed by a direct company announcement, press release, tier-1 media report, specialized semiconductor source, or relevant regional publication.
We excluded semiconductor equipment companies, materials suppliers, EDA and IP-only vendors, packaging and test-only service providers, distributors, electronics manufacturing services, end-product OEMs, broad AI/software companies, chip-design automation tools, and companies where semiconductor devices or foundry services did not appear to be the primary revenue model. Undisclosed-amount rounds were excluded because including them would distort dollar-based metrics such as averages, medians, category shares, and concentration ratios.
The final YTD 2026 dataset contains 32 disclosed deals across 32 unique companies. Every average, median, share, and concentration ratio in this page is computed on that disclosed sample. Privately raised rounds that were never publicly announced are necessarily missing, which is a known limitation of any public-only semiconductor funding tracker.
Related blog posts
- The latest update in the semiconductor industry
- The latest news in the semiconductor industry
- How large is the semiconductor industry really?
- The latest funding news in the semiconductor industry
- The evolution of funding activity in the semiconductor industry
- Which companies have raised the most funding in the semiconductor industry?
- Which companies are the most valued in the semiconductor industry?
