Why Top Contract Manufacturers Are Building Their Own Metal 3D Printers

When buying isn't an option

Norsk Titanium broke the pattern in 2007. Seurat Technologies followed in 2015. VulcanForms in 2015 as well. Freeform in 2019. Divergent Technologies in 2026. Each company looked at the available metal 3D printing systems and concluded the same thing: the commercial options would not get them where they needed to go.

This isn't a case of companies wanting to own their supply chain for PR purposes. The technical and economic arguments are concrete. Existing machines print too slowly, cost too much to operate at scale, or lock users into proprietary material parameters. For contract manufacturers whose entire business model depends on delivering parts faster and cheaper than forging or machining, those constraints are deal-breakers.

Seurat's blank-slate approach

Seurat spun out of Lawrence Livermore National Laboratory with a specific frustration. Founder James DeMuth calculated that printing a 12-meter diameter fusion chamber with conventional laser powder bed fusion would take longer than his remaining lifespan. That wasn't a metaphor. It was a literal calculation that made the existing technology look non-viable for many real-world applications.

The company engineered its own laser technology, light valves, print chambers, and gantry systems. The result, Area Printing, uses pulsed infrared lasers combined with blue light projection to melt material. The optics control cooling rates on a per-pixel basis. Seurat claims this delivers speed and surface finish that conventional LPBF cannot match. The company is targeting 30 kg per hour in a 1.2 meter cube build volume by next year, and 1,700 kg per hour in a 9.6 meter cube by 2030.

Seurat does not sell printers. It sells finished metal parts. DeMuth argues that a single platform based on Area Printing would cost a customer half a billion dollars to buy. His company offers the same capability for a per-part price.

Divergent's end-to-end bet

Divergent took a different path. The company initially used Nikon SLM Solutions machines, even becoming the largest customer of the 12-laser NXG XII 600 platform. But founder Kevin Czinger confirmed earlier this year that all new capacity uses Divergent's own in-house machine, the Divergent Evolutionary Printer. These are integrated into the Divergent Adaptive Production System, or DAPS, which combines AI-driven design, additive manufacturing, and robotic assembly.

The Monolith One, unveiled recently, is designed and built entirely in-house. It is not commercially available. Divergent uses it exclusively for its own contract manufacturing business, producing structures for aerospace, defense, and automotive customers. The machine handles aluminum, nickel, steel, and titanium alloys on the same platform without requiring reconfiguration that would eat into production time.

Divergent's argument is similar to Seurat's: control over the entire stack lets the company move faster than a system integrator ever could. When Saab needed a 15-foot fuselage structure for an autonomous aircraft concept, Divergent designed, printed, and robotically assembled it without any unique tooling or fixtures. The build comprised 26 unique printed parts joined in a single robotic cell.

The impact on machine OEMs

Companies like SLM Solutions, EOS, and Trumpf now compete not only with each other but with their own customers for attention from the world's largest manufacturers. That is an awkward position. When a contract manufacturer builds its own printer, it removes a potential machine sale but gains a reference customer for its parts business.

Some machine OEMs have responded by offering availability-based service agreements and co-development partnerships. Divergent's early relationship with SLM Solutions included beta testing and joint hardware development. That arrangement eventually gave way to full vertical integration, which suggests the OEM partnership model has limits when the contract manufacturer's ambitions grow.

3DEO, a California metal AM company that also developed proprietary technology, filed for insolvency in early August. Its patents and machinery are headed to auction. That outcome is a reminder that building your own printer is a high-stakes gamble. It requires capital, engineering depth, and enough volume to justify the investment. Companies that cannot reach scale may find themselves worse off than if they had simply bought off-the-shelf systems.

Where this is heading

The trend toward vertically integrated metal AM will probably accelerate. As more contract manufacturers reach the scale where machine purchases become a significant line item, the math for in-house development improves. Add in the longer term benefits of owning the intellectual property and the business case becomes harder to ignore.

For end users, this competition between machine OEMs and contract manufacturer-printers should lower costs and improve service levels. Companies like Seurat and Divergent are effectively telling the established players that their machines are not good enough. That kind of feedback from the market usually forces innovation, which benefits everyone.