Nerfstudio

Free plan APILinuxSelf-hostedWebWindows

What Nerfstudio is

Nerfstudio is a free, open-source project with modular tools and an API for creating, training and testing Neural Radiance Fields. Its included methods are Nerfacto, Instant-NGP, NeRF, Mip-NeRF, TensoRF and Splatfacto, its Gaussian Splatting implementation. Command-line tools support data processing, model training and evaluation, video rendering, viewing trained models and exporting them. A web viewer can display training in real time, let users interact with scenes, render custom camera trajectories and create videos from trained models. Share links let others view training progress, but they are publicly accessible and can add latency. Documented capture-to-render workflows use COLMAP, Polycam or Record3D to turn phone video into a 3D render. Installation options include pip, source and Docker; Docker requires NVIDIA GPU drivers capable of CUDA 11.8, and the quickstart requires an NVIDIA video card with CUDA installed. Launched in October 2022, Nerfstudio is developed by Berkeley students and community contributors. Windows installation is less tested and more fragile than Linux.

Who it suits

Nerfstudio suits developers and researchers working with NeRF or Gaussian Splatting workflows who can use its command-line tools and GPU requirements. Linux users may have a smoother documented installation path than Windows users.

What it does well

Free and open-source
Includes six NeRF and Gaussian Splatting methods
Web viewer shows training in real time
Capture workflows include COLMAP, Polycam and Record3D
Offers pip, source and Docker installation options

What to know first

Quickstart requires an NVIDIA video card with CUDA
Docker requires NVIDIA GPU drivers capable of CUDA 11.8
Windows installation is less tested and more fragile
Shared viewer links are public and can add latency

Limits

The quickstart requires an NVIDIA video card with CUDA installed. Docker requires NVIDIA GPU drivers capable of CUDA 11.8. Windows installation is less tested and more fragile than Linux.

Getting started

Choose pip, source, or Docker installation.
For Docker, use NVIDIA GPU drivers capable of CUDA 11.8.
Use the command-line tools to process data and train a model.
Open the web viewer to inspect training or trained scenes.
Use COLMAP, Polycam, or Record3D for the documented phone-video capture workflow.

Verdict

Nerfstudio offers an end-to-end set of tools for training, viewing and exporting NeRF scenes. Its GPU requirements and less-tested Windows installation are important constraints to consider.

Questions

Is Nerfstudio free?

Yes. It has a free plan and is described as an open-source project.

What methods does it include?

Included methods are Nerfacto, Instant-NGP, NeRF, Mip-NeRF, TensoRF, and Splatfacto.

What platforms does it support?

Listed platforms are API, Linux, self-hosted, web, and Windows. Windows installation is less tested and more fragile; Linux is recommended when available.

Does Nerfstudio require an NVIDIA GPU?

The quickstart requires an NVIDIA video card with CUDA installed. Docker requires NVIDIA GPU drivers capable of CUDA 11.8.

Can other people view training progress?

Yes. The viewer supports share URLs, which are publicly accessible and may introduce latency.

What installation options are available?

The documentation provides pip, source, and Docker installation options.

Facts from Nerfstudio's own pages

Capture integrations: The project describes a capture-to-render pipeline using Colmap, Polycam, or Record3D. · github.com, 30 Sept 2026
Data workflows: The project documents full pipeline support with COLMAP, Polycam, or Record3D for turning phone video into a 3D render. · github.com, 29 Sept 2026
Development: Nerfstudio launched as an open-source project by Berkeley students in October 2022 and is developed by Berkeley students and community contributors. · docs.nerf.studio, 30 Sept 2026
Founded: 2022 · nerf.studio, 28 Sept 2026
GPU requirement: The repository quickstart requires an NVIDIA video card with CUDA installed; it says the library was tested with CUDA 11.8. · github.com, 30 Sept 2026
Included Gaussian splatting: Splatfacto is Nerfstudio’s Gaussian Splatting implementation. · docs.nerf.studio, 30 Sept 2026
Included methods: Included methods are Nerfacto, Instant-NGP, NeRF, Mip-NeRF, TensoRF, and Splatfacto. · docs.nerf.studio, 29 Sept 2026
Installation: The documentation provides installation instructions using pip, source code, Pixi, and Docker. · docs.nerf.studio, 30 Sept 2026
License: The project’s GitHub page identifies its license as Apache-2.0. · github.com, 30 Sept 2026
Logging and debugging: The project supports Tensorboard and Wandb logging interfaces, code profiling, and built-in debugging tools. · github.com, 30 Sept 2026
Methods: Included methods include Nerfacto, Instant-NGP, NeRF, Mip-NeRF, TensoRF, and Splatfacto. · docs.nerf.studio, 30 Sept 2026
Platform limits: The documentation says Windows installation is less tested and more fragile, and recommends Linux when available. · docs.nerf.studio, 29 Sept 2026
Platform support: The installation guide says Nerfstudio on Windows is less tested and more fragile, and recommends Linux when available. · docs.nerf.studio, 30 Sept 2026
Project: The documentation describes Nerfstudio as an open-source project launched by Berkeley students in October 2022 and developed by Berkeley students and community contributors. · docs.nerf.studio, 29 Sept 2026
Purpose: Nerfstudio provides a modular API for creating, training, and testing Neural Radiance Fields (NeRFs). · docs.nerf.studio, 30 Sept 2026
Remote viewing: The viewer supports share URLs that let others view training progress, and the documentation says these links are publicly accessible and introduce latency. · docs.nerf.studio, 29 Sept 2026
Support: The project directs users with questions to its Discord and provides documentation, tutorials, and a feedback form. · docs.nerf.studio, 30 Sept 2026
Training tools: Its command line includes tools to process data, train and evaluate models, render videos, view trained NeRFs, and export models. · docs.nerf.studio, 29 Sept 2026
Viewer: Its web-based visualizer can show training in real time, let users interact with scenes, and render custom camera trajectories. · github.com, 30 Sept 2026
Viewer integrations: The web viewer is built using Viser and ThreeJS and packaged as a ReactJS application that connects to a server through a websocket. · docs.nerf.studio, 29 Sept 2026
Web viewer: The viewer displays training in real time and can create content videos from trained models. · docs.nerf.studio, 29 Sept 2026

Where Nerfstudio stands on Freedom251

AI 3D Scanning Software#13 of 24