Xometry launches new AI lead-time prediction and personalized pricing models to improve manufacturing speed, sourcing accuracy and delivery reliability.
Xometry, an AI-driven marketplace connecting buyers and suppliers in custom manufacturing, has introduced new predictive lead-time and dynamic pricing models designed to improve sourcing accuracy and speed across its platform.
The company said the new Enterprise Machining Lead Time Prediction Model and enhanced pricing intelligence expand the capabilities of its Instant Quoting Engine, the core system that connects engineers and procurement teams with a global network of manufacturing partners.
Together, the upgrades aim to deliver more accurate production timelines, faster delivery options and improved operational efficiency for enterprise buyers.
Also Read: The End of the Predictable B2B Buyer Journey
Predicting Manufacturing Lead Times With AI
Xometry’s Instant Quoting Engine uses machine learning models trained on historical production and delivery data generated across its global supplier network. The newly released lead-time model significantly expands the scope of that predictive intelligence.
According to the company, the system now analyzes a training dataset four times larger than previous versions, incorporating new variables such as specialized certifications, additional materials and advanced finishing requirements.
Key improvements include:
- Higher prediction accuracy, with measurable improvements in RMSLE performance compared with earlier models
- Expanded rapid-delivery options, including optimized one-day lead times for a broader set of materials and part geometries
- Improved operational throughput, enabling shorter standard lead-time estimates for customers
The model continuously learns from supplier performance data, allowing the platform to refine delivery forecasts and improve execution reliability over time.
Personalized Pricing for Manufacturing Orders
Alongside lead-time improvements, Xometry has also upgraded the dynamic pricing logic within its quoting system.
Rather than relying on static price tables commonly used in supply chain software, the company’s model analyzes part geometry, quote configurations and a customer’s historical purchasing data to generate a price-response function for each individual quote.
Following testing in late 2025, the personalized pricing models are being rolled out more broadly to U.S. customers during the first quarter of 2026.
The company says the changes are designed to improve both the buyer experience and revenue efficiency across the marketplace.
Also Read: From AI Vision to Retail Personalization at Scale
A Closed-Loop Manufacturing Intelligence System
“These updates represent more than incremental improvements,” said Vaidy Raghavan, Xometry’s chief product and technology officer. “By closing the data loop with our partner network and accelerating model training cycles, we are reducing the time from insight to real-world production impact.”
Xometry’s marketplace integrates quoting, supplier selection, production performance and delivery outcomes into a continuous learning system. Each completed order feeds additional data back into the platform’s AI models, improving future predictions for pricing, delivery and supplier matching.
The company says the approach is designed to address long-standing inefficiencies in manufacturing procurement, where sourcing custom parts has traditionally required manual coordination across fragmented supplier networks.
By embedding predictive intelligence directly into engineers’ workflows, Xometry aims to transform the process into a faster, data-driven system that links digital design with real-world production.
