Xometry, Inc., an international AI-native marketplace that connects buyers and suppliers of customized manufacturing services, on July 30, 2026, announced a series of enhancements to its AI model architecture. As per the company, the new generation of interconnected AI models for customized manufacturing use proprietary data in areas such as manufacturability, supplier capability, geometry, and pricing, as well as production outcomes, so as to help the manufacturing workflow from quotation to production.
As per Xometry, the upgraded models are designed to accelerate and improve the precision of translating engineering data in manufacturing recommendations as well as pricing decisions.
Recommendations for a better manufacturing process
As part of upgrade the AI models for customized manufacturing, Xometry has gone on to improve its context-aware AI process recommender to better comprehend the intended use of the part, its material requirements, and its geometry. The company says the system may deduce application-specific needs, such as higher tolerances and aerospace-grade materials for aerospace parts, so it can make suggestions even for customers having no prior order records.
While uploading a CAD model, the system suggests the most suitable manufacturing process selected from over 20 supported manufacturing technologies. Xometry says clients agree with the AI-generated recommendation over 85% of the time, quite a boost over the old model, especially for first-time users.
Refined cost prediction models
Interestingly, the company has also rolled out a new generation of cost-prediction models that are intended to calculate prices based upon the manufacturing needs of individual components.
The models take into account a broader spectrum of factors, like geometry, material, and surface finish as well as if a component is made as a standalone item or as an element in a bigger production batch, says Xometry. This leads to more precise price estimations.
It is well to be noted that first the new models have been deployed on CNC machining and are trained on the proprietary manufacturing data of Xometry and apparently enhance CNC cost-prediction precision by about 15% using live testing. The models also go on to generate a predetermined level of confidence for each quote prior to returning it to the customer.
Xometry Intends to Roll Out Updated Costing Models to More Manufacturing Processes when it comes to its Instant Quoting Engine – IQE and Increase the Variety of CNC Materials Accessible on the Platform.
Better supplier matching
Notably, the company has also revised its procurement models to enhance assignment of jobs throughout its supplier network.
Xometry utilises data on supplier capability, and its flexible models match manufacturing jobs to the right partners. Criteria for a suitable match include machine size, process capabilities, past performance standards, and on-time delivery.
The company says this method enhances partner matching by providing manufacturers with manufacturing possibilities that are more suitable for their skills compared to general job notifications. Xometry believes that this will boost price responsiveness and make the supplier network as a whole stronger.
According to the Chief Technology Officer at Xometry, Vaidy Raghavan, “Across the millions of parts quoted through Xometry, our AI is constantly learning. Our new models move beyond treating that knowledge as separate skills, where one model recommends how to build something, another prices it, and another finds who should build it. Now, they are an intelligence layer that gets sharper every time a part moves through it. That’s what makes Xometry the digital infrastructure for custom manufacturing.”
He explained, “The data layers are what’s actually deciding how a part gets made, what it costs, and who builds it. And the smarter that data gets, the more directly our customers feel it – more optimal recommendations, accurate pricing, better-fit suppliers – every time they upload a part.”