Poly
Poly is an AI-enabled texture creation tool that lets you quickly generate customized, 8K HD, and seamlessly tile-able textures with up to 32-bit PBR maps using a simple prompt (text and/or image) in seconds. It's perfect for use in 3D applications such as 3D modeling, character design, architecture visualization, game development, AR/VR world-building, and much more. We're thrilled to share the result of our team's research work with the community and hope you will find it useful and fun. Type in a prompt, select a texture material type, and watch as Poly creates a fully-formed 32-bit EXR texture for you. You can use this to play around with Poly's AI, seeing what it is capable of and experimenting with prompting strategies. The dock at the bottom of the screen lets you switch views. You can view your past prompts, view a model in 3D, or view any of the six available physical-based rendering maps.
Learn more
Next3D
Next3D.ai is a browser-based AI 3D creation platform for generating, previewing, editing, exporting, and sharing 3D assets. It helps game developers, 3D artists, designers, e-commerce teams, educators, and prototyping teams turn text prompts or reference images into usable 3D models. The workspace includes an interactive web viewer, model editing and optimization tools, multiple export formats such as GLB, OBJ, STL, and FBX, and shareable links for review. Teams can use it for game props, product visualization, AR/VR experiments, teaching materials, and rapid creative iteration without starting every asset from scratch.
Learn more
Shap-E
This is the official code and model release for Shap-E. Generate 3D objects conditioned on text or images. Sample a 3D model, conditioned on a text prompt, or conditioned on a synthetic view image. To get the best result, you should remove the background from the input image. Load 3D models or a trimesh, and create a batch of multiview renders and a point cloud encode them into a latent and render it back. For this to work, install Blender version 3.3.1 or higher.
Learn more
V2Fun
V2Fun is an all-in-one, browser-based AI 3D creation platform that turns text prompts, reference images, and ordinary videos into production-ready 3D models and animations. Users can generate structurally accurate, perspective-correct concept images, create high-fidelity assets from text or single- and multi-view images, and use built-in prompt optimization to enrich inputs with spatial perspective, PBR materials, and fine texture parameters. Its image-to-3D and text-to-3D engines preserve artistic style, character traits, structural detail, and natural lighting while supporting rapid iteration for characters, scenes, game assets, industrial models, and 3D-printable objects. Smart retopology automatically produces clean, lightweight, editable quad meshes without sacrificing surface detail, while the AI texture generator creates or swaps complete PBR material sets with accurate lighting, bump, gloss, and roughness information.
Learn more
