How the United States and China fund additive manufacturing, 2023-2026. Same industry, two capital stacks, different bets at every layer.
CN rounds 1.82× US median
Across 2023 through May 2026, the additive manufacturing industry in the United States and China drew capital from almost entirely different places, deployed it through different stages, and routed it into different parts of the value chain. The same industry, two capital stacks.
The contrasts to watch are these: where the money comes from, how heavy each round is, what fraction of the cohort gets allocated to physical hardware and materials versus services and software, and which process families attract concentration. Most of these signals hold under three nested cohort definitions (inclusive, AM-ecosystem, strict AM); the one signal that does shift under tightening is the dominant US capital source, and the sensitivity panel below names that explicitly.
US AM funding flows from a mosaic of federal grant programs, corporate strategic capital, and Tier-1 venture firms. China's flows largely from public markets and the "Other" bucket: an opaque mix of provincial industry funds, corporate placements, and small-disclosure rounds.
The dollar weights tell the story most directly. Public markets (IPOs, post-IPO offerings, and private placements treated as quasi-public) put $1.7B into Chinese AM in this window, dominated by the Shanghai STAR Market and Hong Kong listings of equipment makers (Farsoon, Bright Laser, Aidite, Zrapid) and post-IPO placements by listed industrial conglomerates. In the United States the same bucket totals $2.5B, with UL Solutions's 2024 IPO and 2025 secondary offering doing most of the work.
US Federal grant rounds are numerous (64 disclosed transactions) but small in dollars (about $0.28B aggregate). SBIR/STTR Phase I awards typically run in the $200K to $2M range; the high count, low aggregate is the SBIR shape itself, not a US-China gap. China runs analogous early-stage public R&D support through provincial industry funds, national-level guidance funds, and university spin-off vehicles, but those flow on different timelines and report through different channels than the single-round disclosures this report aggregates, so they don't roll up into a single comparable bucket.
The US Strategic/Corporate bucket carries a single load-bearing transaction: Apollo Global Management's $5.2B 2023 acquisition of Arconic, an aluminum supplier with an additive Ampliforge hybrid AM-forging process. Without Arconic, US Strategic/Corporate falls to about $1.6B and the US/CN Strategic gap closes considerably. The hero-anchor section discusses this edge explicitly.
The shape of an individual round is the most stable signal in this comparison. CN median rounds are heavier than US median rounds by 2-3x across all three cohort tiers. US rounds cluster at the small end (SBIR/seed range) and at the very large end (megarounds), with a relatively thin middle. CN rounds cluster squarely in the $5-25M and $25-100M brackets.
The US distribution is visibly bimodal: a $1-5M cluster (driven by SBIR Phase I/II grants and small seed rounds) plus a $25-100M cluster (Series B/C and late-stage corporate). The middle of the US distribution is comparatively thin. CN's distribution by contrast is a single fat band running from $5M through $100M. This is consistent with the Chinese AM cohort being weighted toward later-stage industrial operators backed by industry-fund and public-market capital rather than venture seed-stage formation.
Both cohorts allocate capital across the AM value chain. They do not allocate it the same way.
Combining Hardware and Materials, China funds the physical stack: 88% of the CN cohort sits in those two categories versus 60% of the US cohort. Service and Platform together account for 33% of US funded AM-ecosystem companies and 9% of CN, a roughly four-fold US tilt toward the digital-manufacturing layer (on-demand services, marketplaces, build-process orchestration).
The Hardware count is identical in absolute terms before normalization (104 CN Tier B companies, 77 US Tier B companies; the US's higher overall cohort size means a smaller percentage share). The two cohorts are not racing each other on equipment formation count. They are differing on which layers above hardware get funded enough to count as separate businesses.
At the process-family level (Metal AM, Polymer AM, Ceramics/Composites, Bioprinting, Novel Process), the two cohorts focus differently.
Metal AM is more dominant in the Chinese cohort (56.6% vs 38.1% in the US), consistent with the hero-anchor list being dominated by metal PBF-LB equipment makers and metal-powder producers. Polymer AM is slightly more prevalent in the US (40.5% vs 36%). The cleanest divergence is in Bioprinting, where the US cohort tags at 7.1% versus China's 1.5%, an order-of-magnitude gap. Novel Process and Ceramics/Composites are close in both cohorts.
The fifteen largest disclosed rounds in each country, 2023 through 2026-05-19. Each card carries an explicit AM-relevance label and a one-sentence rationale, so the reader can weigh each megaround on its own terms rather than reading the AMPulse classification as the last word.
The shape of the two columns differs in three visible ways. First, 12 of the 15 CN top rounds are primary-AM companies (pure-play equipment, materials, or service businesses) versus 6 of the 15 US top rounds. Second, 9 CN top rounds are IPOs, post-IPO placements, or private placements versus 3 US top rounds: the dominant CN exit and capital-raise channel is public markets, the dominant US channel is private growth equity plus the occasional acquisition. Third, the largest single US round (Arconic, $5.2B) is a PE buyout of an aluminum supplier rather than an AM-specific raise; the next two US rounds (Coherent, UL Solutions) tag as AM-adjacent at best.
Top 15 megarounds, 2023-2026
Aluminum producer with an additive Ampliforge hybrid AM-forging process and metal powders for aerospace; the 2023 $5.2B figure is Apollo Global private-equity buyout of the entire company, not an AM-specific raise.
Coherent is a global laser and photonics conglomerate; AM is a segment of its industrial-laser business, not the company's primary focus.
Global testing, inspection, and certification firm; AM is one of many practice areas. The 2024 IPO and 2025 secondary offering raised capital for the corporate parent, not an AM-specific program.
Vast is a commercial space station company; metal AM is used in propulsion and structural components but the business is space habitation, not AM supply.
NatureWorks manufactures Ingeo PLA biopolymer across packaging, fibers, and 3D printing; AM is one application sector, not the company's primary market.
Defense technology company building unmanned systems and precision-strike munitions; uses additive manufacturing for rapid drone production but the business model is autonomous-systems integration.
Divergent's entire platform is metal AM (DAPS) for automotive vehicle structures; AM is the product.
Pure-play metal additive manufacturing service for aerospace, medical, and defense; AM is the entire business.
AM hardware OEM across metal binder jet, MEX, and ceramics; pure-play.
Medical-device company built on 3D-printed bioabsorbable orthobiologics (Tapestry RC implant family); AM is the manufacturing core.
Lunar surface construction via in-situ resource utilization with 3D concrete printing of regolith. AM is the build method but the company is a space-infrastructure bet, not an AM-supply business.
Rocket propulsion systems for commercial and defense use, with metal AM as the manufacturing approach for engine components; AM is integral to production but the business is rocket engines, not AM supply.
Vertically integrated supplier of refractory metal powders (tungsten, molybdenum) and precision-engineered AM components, specializing in binder jetting and DED for aerospace and defense.
Pure-play 3D-printed personalized orthopedic implant maker (Metal AM laser powder-bed fusion) with AI-driven implant design and osseointegrative materials; AM is the manufacturing and product core.
Atomic Layer Deposition (ALD) nano-coating technology applied across battery materials, semiconductors, and AM powders; AM is one of several markets the company serves, not the core business.
Top 15 megarounds, 2023-2026
Pure-play metal AM equipment, parts, and services for aerospace and energy customers; listed on Shanghai STAR Market.
AM-aligned inspection and metrology technology, including CT scanning and process monitoring for additive parts.
STEM education robotics and 3D-printer maker for classrooms (mBot family, Laserbox); the AM angle is consumer/education hardware, not industrial AM supply.
Industrial laser equipment conglomerate (Han's Laser group context) with metal AM (DED-LB and PBF-LB) as one product line; AM is one of several laser-application businesses.
Multi-process AM service bureau (metal PBF-LB and polymer MEX/PBF/VPP-SLA) listed on Shenzhen exchange; AM service is the business.
CNMNC/OTIC is the AM-focused subsidiary group of state-owned China National Machinery Industry Corp covering metal PBF-LB equipment; AM is the entity's product.
Pure-play industrial AM equipment maker (metal and polymer laser PBF) listed on Shanghai STAR Market in 2023; AM is the entire business.
Dental ceramics and zirconia materials producer with AM (VPP, DLP, SLA) for dental restorations; AM is the manufacturing core.
Laser-welding and battery-equipment manufacturer; AM (polymer) is one product line among industrial laser systems for new-energy manufacturing.
Industrial 3D-printer maker across polymer SLA, polymer PBF, and metal PBF-LB; pure-play AM hardware.
AM platform / marketplace integrating printing service with software workflow; AM-centric platform.
Metal AM equipment maker (Raycham brand) covering PBF-LB and DED systems for industrial users.
Specialty alloy and metal powder maker for additive manufacturing; AM is the materials focus.
Full-stack digital 3D printing for dental and industrial markets: hardware, software, biocompatible resin materials. AM is the integrated business.
AI platform generating 3D models from text or image prompts; positions at the design / pre-processing front-end of the additive manufacturing workflow, displacing CAD as the input layer for 3D-printable geometry. AM-native design software.
Eight findings re-run under Tier A (inclusive), Tier B (default), and Tier C (strict AM tagging). Seven of the eight hold direction across all three tiers. The one row that does not hold (the top US capital source) is a finding about which bucket dominates, not about whether the US-China contrast exists; it indicates that the US capital-source picture is itself cohort-sensitive in a way the CN picture is not. The magnitudes change: as the cohort tightens, the absolute cohort sizes converge (US 338 vs CN 199 at Tier A becomes US 137 vs CN 120 at Tier C) and the US-to-CN disclosed-capital ratio compresses from about 5.9x at Tier A to about 3.0x at Tier C. The median-round size and disclosure-rate gaps are nearly identical across tiers.
| Finding | Tier A (inclusive) | Tier B (default) | Tier C (strict AM) | Holds? | |||
|---|---|---|---|---|---|---|---|
| US | CN | US | CN | US | CN | ||
| Funded companies (US vs CN) | 338 | 199 | 168 | 136 | 137 | 120 | ✓ |
| Disclosed capital (US vs CN) | $37.0B | $6.3B | $13.0B | $3.2B | $9.1B | $3.0B | ✓ |
| Median round size (CN > US?) | $5.1M | $9.0M | $5.0M | $9.0M | $4.5M | $10.0M | ✓ |
| Disclosure rate (rounds with $$ / total) | 73.6% | 50.2% | 68.9% | 50.2% | 66.1% | 49.4% | ✓ |
| Top capital source by disclosed $$ (US) | Tier-1 VC | Strategic / Corp | Strategic / Corp | ✗ | |||
| Top capital source by disclosed $$ (CN) | Public Markets | Public Markets | Public Markets | ✓ | |||
| Hardware+Materials share (the "physical stack") | 50.9% | 77.9% | 60.1% | 87.5% | 64.3% | 90.8% | ✓ |
| Metal AM share | 21% | 39.7% | 38.1% | 56.6% | 33.6% | 60% | ✓ |
Findings re-run under three nested cohort definitions. "Holds" means the sign of the US-CN difference is consistent across all three tiers (i.e. no tier inverts the direction of the finding).
The disclosed-capital ratio compresses but does not invert as the cohort tightens. Under Tier A the US lead is $37.0B vs $6.3B (about 5.9x); under Tier C it is $9.1B vs $3.0B (about 3.0x). The ratio is sensitive to whether AM-adjacent megarounds (Arconic, Coherent, UL Solutions) are counted on the US side. The counter-thesis section below addresses this directly. Of the four structural findings, three (CN's median-round size advantage, CN's heavier physical-stack share, CN's heavier Metal-AM tilt) hold across all three tiers; the fourth (the top US capital source by disclosed dollars) shifts from Tier-1 VC at Tier A to Strategic / Corp at Tier B and Tier C, which is consistent with the report's view that the US picture is more sensitive to where the cohort boundary is drawn than the CN picture.
A determined skeptic could attack the headline three ways. We name each, then answer.
1. CN disclosure rate is low and the absolute capital numbers cannot be trusted. Fair, and this report does not lean on absolute capital totals as primary evidence. The CN cohort's round disclosure rate is 50.2% versus 68.9% for the US, so the CN $3.2B Tier B figure is incomplete and approximate. Disclosed totals here count only rounds with a parsable, currency-resolved USD figure; trade-press estimates and soft scale-language are excluded so the headline number is conservatively counted, not generously imputed. Even doubling the CN disclosure base would not invert the direction. The report's structural findings (which sources, which stages, which layers, which processes) do not depend on the absolute total.
2. AMPulse classification leakage may inflate one side's numbers more than the other's. Also partly true, and the report addresses this directly via the AM-relevance labels on hero anchors and the Tier A/B/C sensitivity cuts. Of the US top 15 megarounds, 9 carry adjacent or edge labels (Arconic, Synopsys, Coherent, UL Solutions x2, Vast x2, NatureWorks, AEVEX, Starcloud), summing to about $8.7B. Of the CN top 15, 3 carry adjacent or edge labels (GigaDevice, Lead Intelligent, Makeblock, HGLASER, Hymson), summing to about $0.51B. Excluding adj/edge rounds on both sides (a top-15-only scrub, not a full strict-AM re-cohort like Tier C) drops the visible US Tier B disclosed total to about $4.3B and the CN total to about $2.7B. The US-CN gap survives the symmetric scrubbing at approximately 1.57x. The report's headline framing leans on the median-round, stage-mix, and category-mix findings, which the sensitivity check above shows are insensitive to whether Synopsys or Makeblock is counted.
3. The snapshot is too narrow. A different 36-month window would tell a different story. The cohort window is 2023-01 through 2026-05-19, roughly 41 months. Capital cycles in private markets typically run on 3-7 year horizons, so this window captures roughly half a cycle. A 2020-2022 window would weight earlier-stage formation more heavily (and would show the post-2021 SPAC wave skewing US totals). A 2024-2026-only window would emphasize the present pattern. We do not claim this report describes "the structural state" of US-CN AM funding; we claim it describes the shape of the past three-plus years and the major rounds within them. Subsequent reports may re-run the same lens against a refreshed window.
What this report does not attempt:
Source. AMPulse's tracked-company database, US and China subset, with funding-round history scoped to 2023-01-01 through 2026-05-19. The US-CN slice covers approximately 2,300 companies of which 537 have at least one round in the window.
Snapshot integrity. All numerical claims in the body are computed from a single snapshot extract. The data files behind this report are frozen as of the snapshot date and versioned alongside the report so subsequent re-extracts do not silently mutate the values cited here.
Corrections policy. Pre-publication patches modify the underlying data files in place and are recorded in revision notes. Post-publication corrections (none at time of writing) would be issued as a companion file with explicit change rationale, leaving the original artifact intact for citation continuity.
