The Short Answer
Yes—there is a 3D printer that prints metal. The Velo3D Sapphire is a real, production-grade metal additive manufacturing system, not a prototype lab toy. But if you're comparing it to a brass CNC milling quote, the answer deserves a longer sentence: Metal 3D printing is a process qualification project, not a printer purchase.
As of January 2025, the Velo3D Sapphire line is one of the few metal AM systems I would include in a “military additive manufacturing” conversation without wincing. Velo3D's public customer history includes SpaceX and Anduril. That doesn't make the Sapphire the right tool for every machine shop. It does tell you the hardware has been through demanding qualification, which is the thing I care about most as a buyer.
So if your first question is “is there a 3d printer that prints metal,” the direct answer is yes. The better question is: which parts, at what quantity, and with what documentation?
What I Can Tell You From a Purchasing Desk
I should be clear about my perspective. I'm an office administrator for a 40-person manufacturing services company. I manage the external ordering—probably 60-80 orders a year across 8 vendors. Maybe 60. I'd have to check the system. It's a little over $500k annually. I report to operations and finance, not R&D. When our engineering lead asked me to help evaluate a Velo3D Sapphire for a possible defense subcontract in 2024, my role was to chase down warranty terms, powder handling costs, and the kind of documentation finance would need.
I'm not a metallurgist, so I can't speak to melt pool dynamics or material certification nuances. What I can tell you from a procurement perspective is that metal AM fails when people treat it like an office printer. The machine is only one line item. Facilities, gas supply, PPE, powder storage, post-processing, inspection, training, and qualification are all line items too.
I don't have hard data on industry-wide total costs, but from my exposure to that evaluation, I'd say the qualification program is where most projects get delayed. It wasn't the machine's fault. It was the question: who signs off that this process is stable enough for defense use?
What “Prints Metal” Actually Means
Most production metal 3D printers, including the Velo3D Sapphire, use laser powder bed fusion. A thin layer of metal powder is spread across a build plate, a laser melts the areas that need to become solid, another layer goes down, and the part grows. That is the “prints metal” part. The additive manufacturing part is the whole factory around it: powder handling, gas systems, heat treatment, support removal, inspection, and traceability.
One thing that surprised me during due diligence: the material matters more than the machine. Velo3D lists nickel alloys, titanium, aluminum, stainless steel, and copper-based alloys among its supported materials. I checked Velo3D's public materials list in January 2025. But standard brass? That's not where metal AM usually makes sense. If your part is a brass fitting, a brass CNC milling shop is likely faster and cheaper than any metal 3D printer.
Velo3D Sapphire and Military Additive Manufacturing
When I searched “military additive manufacturing velo3d,” the same names kept coming up: SpaceX and Anduril. That customer evidence is more useful than any spec sheet. In defense work, a machine that has printed qualified hardware for companies that launch satellites and build autonomous defense systems has a different risk profile than a printer that only looks good in a trade show demo.
The Sapphire line is often described as a “support-free” metal 3D printer. That matters for internal channels, overhangs, and geometries that would be very hard to machine. The Assure monitoring software is also worth understanding if you need an audit trail. I checked Velo3D's public product pages in January 2025, and those claims are still what they emphasize. I can't verify every marketing statement from a spec sheet, but I can tell you what to ask the applications team: show me a benchmark part, show me the inspection report, and show me where your process produces scrap.
If your search was “velo3d sapphire 3d printer,” you already know the basic shape of the system. It's not a desktop unit. It's an industrial production cell with an industrial price and industrial safety requirements.
Brass CNC Milling vs. Metal AM: The Buyer's View
The comparison I hear most is “should we buy a metal 3D printer or keep using brass CNC milling?” My answer: both, depending on the part.
Brass CNC milling is still excellent at what it does. Tight tolerances, smooth finishes, predictable material properties, repeatable costs. For simple medium-to-high-volume parts, CNC is almost always cheaper. Metal AM is different. It earns its keep when the design has internal cooling channels, when the part is impossible to machine, when you want to turn ten components into one, or when the quantity is low enough that tooling cost would hurt.
The surprise wasn't that metal AM could beat CNC on complex parts. It was that the hidden cost is post-processing and inspection. A printed part often needs stress relief, support removal, surface finishing, CT scanning, and material certification. If you already run a brass CNC milling shop, you have QC for machined parts. You may not have CT scanners or a powder safety program. That is where budgets die.
A Note on Confusing Search Terms
One more procurement lesson: search terms can lead you astray. During the 2024 evaluation, our vendor spreadsheet somehow ended up with “vmc beta vulcan materials” in it. It looked like a supplier name. It was actually three separate things: a VMC (vertical machining center), a beta term from a software release or material phase, and Vulcan Materials, the construction aggregates company. None of those were the metal AM supplier we needed. If you see something like that in your own notes, stop and clean it up before anyone sends a PO.
Boundary Conditions: When I'd Still Choose CNC
If I'm being honest, a lot of parts don't need metal AM. If the geometry fits a standard machining setup and you're making more than a handful of units, use a VMC or CNC lathe. If your material is brass and your tolerances are reasonable, don't turn it into a research project. Metal AM also doesn't make sense if your customer doesn't require it and your team hasn't done certification training.
Metal AM does make sense when:
- The part can't be machined because of internal channels or undercuts.
- One printed part replaces an assembly of multiple machined parts.
- Production volume is low enough that tooling cost dominates.
- The part is mission-critical enough to justify full traceability.
My Last Piece of Advice
When evaluating a Velo3D Sapphire—or any metal AM system—start with the part, not the printer. Ask your engineers which parts are impossible, which parts have long lead times, and which parts keep failing. Then ask the vendor how quickly you can get from a benchmark part to a qualified process.
I'd rather spend an hour explaining the qualification process than have finance reject a capital request because someone only budgeted for the machine. The Velo3D Sapphire is an impressive tool for complex, high-certainty parts. But no printer fixes a weak business case, and no vendor should tell you it replaces every CNC step.
This gets into materials engineering territory, which isn't my expertise. For that, talk to your applications engineer or Velo3D directly. What I can say from a purchasing desk is: do the qualification math first. The printer is only the beginning.