nerfstudio-project/nerfstudio

Health v2 · maintenance only

Adoption not part of the score

11934 stars · 1662 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Nerfstudio is a Python library and CLI toolkit providing a simple, modular API for creating, training, and testing Neural Radiance Fields (NeRFs) and Gaussian Splatting models. It includes a web viewer, data parsers, multiple state-of-the-art methods (Nerfacto, Splatfacto, Instant-NGP, TensoRF, etc.), and tools for exporting geometry to engines like Blender, Maya, and Unreal.

Use cases

• train a NeRF model from photos of a scene
• reconstruct 3D scenes from images with gaussian splatting
• render novel views of a captured scene
• export a NeRF mesh to Blender or Unreal Engine
• interactively view and debug NeRF training in a browser
• implement and benchmark a new NeRF research method
• process video or images into a training dataset with COLMAP

When to choose

• you want an end-to-end, plug-and-play pipeline for NeRF or Gaussian Splatting
• you need a real-time web viewer to inspect training progress
• you want to prototype or extend NeRF methods with a modular API
• you need to export trained radiance fields as meshes or point clouds for VFX pipelines

When to avoid

• you need a lightweight inference-only viewer without training
• your project has no GPU available, since training is GPU-intensive
• you need a general-purpose photogrammetry tool producing textured meshes without neural methods
• you require a non-Python or embedded deployment environment

Facets

library · maturity active

machine-learning deep-learning computer-vision graphics cli developer-tools computer-vision machine-learning graphics deep-learning python cli nerf gaussian-splatting 3d-reconstruction photogrammetry pytorch 3d-graphics viewer novel-view-synthesis linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="nerfstudio-project/nerfstudio")