For now, MOI Composites' robotic large-format process runs on a single material. The company's materials page lists one feedstock for its SFM process: sG A1134, a vinyl ester paste loaded with short glass fibers. That is set to change. According to VoxelMatters, the Italian robotic large-format additive manufacturing (LFAM) specialist will preview four new short-fiber thermoset composite feedstocks at Formnext 2026. Each one targets a different set of applications. MOI will also show a test of reinforcement recovered from ground-up composite parts.
If you mostly print PLA and PETG on a desktop machine, this may sound remote. It is still a useful look at where structural, large-scale additive manufacturing is going. In that world, the open questions are less about layer lines and more about glass transition temperatures, fire codes and what to do with scrap.
The Baseline: sG A1134
The four new materials are easier to judge against what MOI offers today. MOI describes sG A1134 as a vinyl ester matrix containing up to 45% short glass fibers by weight. It is a composite paste rather than a filament, and the company says it is engineered for fast processing and for machining after printing. MOI credits the high filler content with raising stiffness, heat deflection temperature (HDT) and fire performance. The company also cites strong chemical resistance.
The key difference from most material extrusion that makers know is the thermoset chemistry. A thermoplastic like PETG or nylon can be melted again. A vinyl ester cures into a cross-linked network that will not flow when reheated. You give up reprocessability and get better thermal stability and chemical resistance. That trade-off explains why thermoset composites keep showing up in tooling, marine and transport work.
Four Feedstocks, Four Jobs
VoxelMatters reports that each new feedstock is aimed at a specific use case. None of them is pitched as a general-purpose replacement.
- High Tg. This one is aimed at autoclave lamination and hot tooling. A higher glass transition temperature means the printed part can keep its shape at the elevated temperatures used to cure laminated composites. That matters when the printed part is the mold.
- Low Density. This one targets marine, architecture and design. In those fields a lighter part is easier to handle, ship and install, and large panels or forms can become very heavy.
- Tough. This one is intended for end-use parts that have to withstand impact and wear, as opposed to tooling or prototypes.
- Flame-retardant. This one is described as meeting the requirements of transport industries such as rail. In those sectors fire behavior is a requirement, not just a nice extra.
Taken together, the list shows MOI dividing its market by application rather than offering one material for everything. Hot tooling, lightweight architectural forms, impact-resistant end-use parts and fire-rated transport components each have their own qualification hurdles. One paste is unlikely to clear all of them. sG A1134 already claims good fire performance from its filler loading. A dedicated flame-retardant grade points toward the more formal requirements of rail and transport customers.
The VoxelMatters report gives no detailed property data for the new grades. Glass transition values, densities, impact figures and fire-test classifications are all missing, so the names are the only specifics available for now. Readers should treat these as previews, not as datasheets.
Closing the Loop on Composite Scrap
The recycling test may be the most interesting item on MOI's list. VoxelMatters reports that MOI is reintroducing reinforcement recovered from a ground composite part into its material stream. Giulia Tarricone of MOI's Materials Development Team said that in tests to date the mechanical properties have remained "broadly comparable" to those of the reference material.
Recycling is a known weak spot for thermoset composites. A cured vinyl ester part cannot be melted down and re-extruded the way a thermoplastic print can. Grinding it up and using the recovered material as reinforcement is one of the more practical ways to keep composite scrap out of landfill. "Broadly comparable" is a hedge, not a spec sheet. Still, it suggests the recovered material is not simply being used as cheap filler.
The report does not say what fraction of the reinforcement was recovered material, or which specific properties were compared. Those details will decide whether this stays a demonstration or becomes a product.
The Hardware Behind It
The materials are designed for hardware MOI has been building out since 2025. VoxelMatters says the company launched its SFM-based Hybrid Fabrication Platform (HFP) Series that year. It is a production cell with a six- or eight-axis industrial robot, the proprietary S18 toolhead and precision milling. The printhead produces near-net-shape structures, which are then finished by subtractive milling. That fits a material sold for fast processing and post-print machining.
That hybrid workflow is common in large-format composites. At these sizes, printed surface finish is rarely good enough for tooling or mating faces. Planning for machining from the start is more realistic than trying to print finished surfaces.
VoxelMatters also describes "Belotti Powered by Moi," a hardware integration initiative with CNC machining company Belotti SpA. Quoting Belotti, it is a hybrid platform combining additive deposition of thermoset-matrix composites with high-precision CNC milling in a single production environment. The report says Belotti's CNC machining centers can integrate MOI's SFM technology, with no need to reposition or handle parts between the additive and subtractive steps. It also says MOI's wider development program, aimed at an integrated LFAM ecosystem of materials, digital design and robotics, is supported by the European Innovation Council (EIC) Accelerator. Formnext runs November 17 to 20 in Frankfurt, and MOI will be in Hall 12.1, Booth 121-D19.
What It Means for Makers
Most readers will not be buying a robotic thermoset cell. The announcement still matters for a few reasons.
First, it shows how industrial material extrusion is maturing. Machine builders are moving from "here is our material" to "here is the right material for your application." Desktop filament has already gone the same way: generic PLA gave way to specialized high-temperature, flame-rated, low-density and toughened grades. MOI is following that pattern with thermoset pastes, at a much larger scale.
Second, a High Tg grade for hot tooling and autoclave lamination is relevant to anyone who builds molds. Small shops that make carbon or glass layups know that getting a mold to survive cure temperatures is often the hardest part. When LFAM vendors build grades specifically for that job, the economics of printed tooling move a little further toward printing instead of machining from billet.
Third, the recycled-reinforcement test addresses a real weakness of thermoset composites. Re-grinding failed prints is an established practice for thermoplastics. Thermosets have never had an easy equivalent. If MOI's "broadly comparable" result holds up with published numbers, it would help counter one of the strongest objections to printing large structures in non-meltable resins.
The bottom line: this is a preview, not a launch. No specifications, pricing or availability dates are in the report. Watch for MOI to publish property data for the four grades and the recycled blend at or after Formnext. Until then, the most solid information is what is already on MOI's own materials page.
