TripoSR · Global AI Product Index

Overview

TripoSR is a fast feed-forward 3D generative model for reconstructing an object from a single image. Weights are on Hugging Face (stabilityai/TripoSR) and code is on GitHub (VAST-AI-Research/TripoSR) under the MIT license. It is an open research model you run yourself, not a hosted studio app.

Key features

Single-image to 3D reconstruction
Feed-forward (fast) inference
Open weights on Hugging Face
MIT-licensed code on GitHub

Use cases

Reconstructing a mesh from a product photo
Research on fast feed-forward 3D models
Self-hosted image-to-3D prototypes

Pricing

TripoSR weights and code are free under MIT; you pay only for the hardware used to run them.

Frequently asked questions

What is TripoSR?

TripoSR is a fast open model that reconstructs a 3D object from a single image, released with Stability AI and VAST.

Is TripoSR the same as Tripo AI?

No. Tripo AI is a hosted 3D generator. TripoSR is an open feed-forward model on Hugging Face and GitHub.

What is the license?

The model is released under the MIT license.

Similar products

3D AI Studio — Web platform that generates 3D models from text or images using multiple AI engines, with mesh cleanup and exports.
Hunyuan3D — Open Tencent Hunyuan diffusion models for high-resolution textured 3D asset generation from images or text.
Kaedim — AI 3D production service that turns sketches and photos into 3D assets for studios and brands.
Meshy — AI tool that generates 3D models and textures from text prompts and images.
Rodin — 3D generative model from Deemos that creates production-ready 3D assets from text and images.
Skybox AI — AI service from Blockade Labs that generates 360-degree panoramic skybox scenes from text prompts.

All TripoSR alternatives → · All 3D & Spatial products →

Sources

This record was last reviewed on 2026-09-10.

Machine-readable record: /api/products/triposr.json · Spot an error? Suggest a correction