The short URL: https://ask.yugatech.com/9lhk

If you’ve been active in the 3D community, you’ve probably seen their advertisements all around social media.

Meshy has announced a nearly USD 400-million Series B round at a reported USD 1.5-billion valuation, alongside new tools that turn text, images, sketches, and conversations into 3D models for games and 3D printing. The funding and valuation figures come from Meshy’s announcement and have not been independently verified.

The company’s Meshy 3D Agent is designed to handle more than simple text-to-image generation. Meshy says users can describe an object, upload a photo, provide a sketch, or continue a conversation before exporting the resulting model in formats including STL, the file type commonly used by 3D-printing slicers.

That workflow could reduce the amount of traditional computer-aided design work needed before printing. It does not remove the preparation stage, however. Users still need to inspect the model, repair possible geometry problems, choose an orientation, set the scale, and prepare the file for a particular printer and material.

Meshy has also introduced Auto Split, a tool intended to break a larger model into printable parts. The company says Auto Split can repair surfaces, create closed solids, and arrange the pieces for printing before reassembly. A closed solid is a model with a complete outer shell, which is important because slicers need to understand the object’s boundaries before generating toolpaths.

A model that looks correct on screen is not automatically ready for a real printer. Parts may be too large for the build plate, contain non-manifold surfaces, or have overhangs that require support material. Auto Split is intended to address part of that preparation process, but Meshy’s performance claims remain company-reported and should not be treated as a guarantee for every model.

The company also announced Smart Topology, which is intended to improve the structure of generated models. Topology refers to how the surfaces and polygons are arranged. Better topology can make a model easier to edit, animate, or process, although the result still depends on the object and the intended use.

Meshy said its platform had more than 12 million registered users and over 100 million models created as of July 2026. Those figures are also company-reported. The announcement positions the new agent and related tools as part of a broader effort to connect generative AI with usable 3D assets rather than stopping at a visual preview.

AI-generated models could help students, cosplay makers, toy designers, game developers, and small businesses produce custom objects without first mastering advanced CAD software. The important questions remain model accuracy, copyright ownership, printability, and the cost of generating higher-quality assets.

The Meshy 3D Agent is described as available to registered users, but anyone planning to sell printed products should still review the platform’s terms, inspect every model, and confirm that the source material can be used commercially.