VGGT-1B (Visual Geometry Grounded Transformer) is a model from Meta AI and University of Oxford researchers. It predicts camera intrinsics/extrinsics, depth maps, point maps, and 3D tracks directly from images in seconds. Released as open weights on Hugging Face, it targets computer vision researchers and developers working on 3D reconstruction and geometry tasks.
VGGT 1B sits in PulseGate's Other AI category. It focuses on inferring complete 3D scene geometry including depth, camera parameters, and point tracks from one or multiple images. It is built as an open-source project for developers. VGGT 1B is open source under the Open Source license. VGGT 1B is available on the web and API.
It is developed by Meta AI, and it first shipped in 2025. The project is developed in the open on GitHub with 13.9k stars and 2 commits in the last 90 days. Among its 3 catalogued features are 3D Reconstruction, Depth Estimation, and Camera Parameter Prediction.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do