On the recordRead 11 Oct 202622 facts

Mitsuba 3

What it is

Mitsuba 3 is a research-oriented system for forward and inverse light-transport simulation. It supports monochrome, RGB, and spectral rendering, with optional polarization tracking, and computes derivatives of simulations with respect to inputs such as camera pose, geometry, materials, textures, and volumes. Dr.Jit compiles rendering code for CPUs through LLVM and NVIDIA GPUs through CUDA and OptiX. Its Python integration supports developing materials, textures, and rendering algorithms, and the documentation includes a PyTorch interoperability tutorial. Installation is recommended with pip and requires Python 3.10 or later. GPU computation optionally requires NVIDIA driver 535 or later, while parallel CPU computation optionally requires LLVM 18 or later. The project says it has been tested on Linux x86_64, macOS aarch64 and x86_64, and Windows x86_64. The Python package includes thirteen variants by default; additional variants require compiling a custom Dr.Jit version using CMake. Tutorials, notebooks, how-to guides, and API and plugin references are provided. Mitsuba 3 is written in portable C++17 on Dr.Jit, with plugins spanning materials, light sources, and rendering algorithms.

For and against

For

Supports monochrome, RGB, and spectral rendering.
Computes derivatives with respect to scene inputs.
Renders on CPUs and NVIDIA GPUs.
Includes Python integration and a PyTorch tutorial.

Against

Requires Python 3.10 or later.
Only thirteen variants are included in the Python package by default.
Extra variants require compiling a custom Dr.Jit version.

The verdict

Mitsuba 3 offers configurable light-transport simulation and differentiable rendering for research workflows. Check its Python requirement and optional GPU or CPU dependencies before installation.

Who it suits

Mitsuba 3 is intended for research and experimentation in computer graphics and other disciplines. Its Python workflow and tutorials cover applications including inverse rendering and shape optimization.

Plans and prices

Free

Python package includes 13 variants by default · Python >= 3.10 · GPU use optionally requires Nvidia driver >= 535 · CPU vectorization optionally requires LLVM >= 18

Read 11 Oct 2026 · github.com

Every plan, in full

On the record

Architecture: It is written in portable C++17 on top of the Dr.Jit just-in-time compiler.mitsuba-renderer.org · read 11 Oct 2026
Audience: The project describes itself as research-oriented and provides tutorials for applications including inverse rendering and shape optimization.mitsuba.readthedocs.io · read 11 Oct 2026
Color and light: It supports monochrome, RGB, and spectral rendering variants, with optional polarization effects.github.com · read 11 Oct 2026
Compatibility: Mitsuba 3 strives to retain scene compatibility with Mitsuba 0.6 and Mitsuba 2.mitsuba.readthedocs.io · read 11 Oct 2026
Developer: The project says it was created by Wenzel Jakob and developed at EPFL in Switzerland.github.com · read 11 Oct 2026
Differentiable rendering: It computes derivatives of simulations with respect to inputs such as camera pose, geometry, materials, textures, and volumes.github.com · read 11 Oct 2026
Installation: The recommended installation method is pip, which also installs Dr.Jit if it is not already available.mitsuba.readthedocs.io · read 11 Oct 2026
Integration: The documentation includes a tutorial on interoperability between PyTorch and Mitsuba.mitsuba.readthedocs.io · read 11 Oct 2026
Integrations: The documentation includes a tutorial on interoperability between PyTorch and Mitsuba.mitsuba.readthedocs.io · read 11 Oct 2026
Intended users: The project describes itself as research-oriented and says its Python workflow enables experimentation in computer graphics and other disciplines.github.com · read 11 Oct 2026
Learning resources: The documentation provides tutorials, notebooks, how-to guides, and API and plugin references.mitsuba.readthedocs.io · read 11 Oct 2026
License: The repository license permits redistribution and use in source and binary forms subject to its listed conditions.github.com · read 11 Oct 2026
Limits: The Python package includes thirteen variants by default; additional variants require compiling a custom version of Dr.Jit using CMake.github.com · read 11 Oct 2026
Performance: Dr.Jit compiles rendering code for CPUs through LLVM and for NVIDIA GPUs through CUDA and OptiX.github.com · read 11 Oct 2026
Plugin system: Its core libraries and plugins cover functionality from materials and light sources to complete rendering algorithms.mitsuba-renderer.org · read 11 Oct 2026
Purpose: Mitsuba 3 is a research-oriented renderer for forward and inverse light-transport simulation.mitsuba-renderer.org · read 11 Oct 2026
Python: Mitsuba 3 integrates with Python, where users can develop materials, textures, and rendering algorithms.github.com · read 11 Oct 2026
Rendering variants: It supports monochrome, RGB, and spectral rendering, with optional polarization tracking.github.com · read 11 Oct 2026

Alternatives to Mitsuba 3

TerragenWindowsMac+2 D5 RenderWebWindows OctaneRenderiPhone & iPadWindows+2 Light Tracer RenderAndroidiPhone & iPad+3 RenderManWindowsMac+1 LuxCoreRenderWindowsMac+4 Maxwell RenderWebWindows+1 Thea RenderWindowsMac+1

Every alternative, ranked