nvpro-samples/vk_gltf_renderer

Vulkan ray tracing glTF 2.0 renderer — RTX path tracer with PBR materials, AI denoising (DLSS, OptiX), and scene editing. NVIDIA open-source sample.

What it solves

This project provides a production-quality Vulkan RTX path tracer and a reference implementation for glTF 2.0 PBR (Physically Based Rendering) materials. It addresses the need for graphics developers to have a high-fidelity, extensible tool for studying, profiling, and authoring glTF scenes with physically accurate lighting and materials.

How it works

The renderer uses a Monte Carlo path tracer with Next Event Estimation (NEE) and Multiple Importance Sampling (MIS) to evaluate glTF PBR materials. It is built in C++ using Slang shaders and leverages NVIDIA RTX hardware for ray tracing. To maintain interactive rates, it integrates AI-driven denoising via DLSS Ray Reconstruction and the OptiX AI Denoiser. It also includes a rasterizer preview for fast iteration during scene editing.

Who it’s for

It is designed for graphics developers and technical artists who need a reference for implementing PBR materials, a tool for non-destructive glTF scene authoring, or a high-performance Vulkan ray tracing implementation to study and extend.

Highlights

  • glTF Scene Editor: Non-destructive editing of hierarchies, materials, and lights with the ability to save changes back to glTF.
  • Extensive glTF Support: Support for 33+ glTF extensions, including anisotropy, clearcoat, transmission, and volume.
  • AI-Enhanced Rendering: Integration of DLSS Ray Reconstruction and OptiX AI Denoiser for clean, interactive images.
  • Agentic AI Generation: An optional ComfyUI bridge allowing users to beautify renders or generate 360° HDRI environments from text prompts.
  • Complex Scene Composition: Preview support for glTF 2.1 external assets, allowing for nested references and shared geometry.
  • Developer Tooling: Includes a GPU profiler, memory tracker, and shader hot-reload capabilities.

Related

  • Project
    NVIDIA-RTX/NRD

    A spatio-temporal denoising library for real-time path tracing that resolves noise on opaque surfaces using G-buffer guides across D3D12, Vulkan, and D3D11.

  • Project
    NVIDIA-RTX/RTXNS

    A developer toolkit for integrating machine learning into graphics applications, enabling the use of neural networks to represent shaders and textures.

  • Project
    apistol78/traktor

    A high-performance open-source 3D game engine written in C++ featuring advanced Vulkan rendering, hardware ray tracing, and built-in AI integration via an MCP server.

  • Project
    Devsh-Graphics-Programming/Nabla

    A low-level GPU programming framework based on Vulkan and SPIR-V that provides a curated feature profile and a shared HLSL/C++ library for cross-platform graphics development.

  • Project
    NVIDIA-RTX/RTXNTC

    A neural texture compression SDK that compresses multiple PBR texture channels into a single neural set to reduce VRAM and bandwidth usage in 3D rendering.