OpenDLSS: A Vulkan Reimplementation of Nvidia's DLSS 5 Neural Rendering Network

In one line

A Vulkan reimplementation of NVIDIA’s DLSS 5 Neural Rendering network, bit-exact against the original. - maanHimself/OpenDLSS-NR

Opening excerpt

The same 71-block Swin / ViT network as DLSS-NR build 310.8.0, running FP8 on the tensor cores. The intermediates match too, not just the final image: all 75 block boundaries, byte for byte.

A U-net of shifted-window transformer blocks with a global ViT at the bottom: 71 blocks over six pooling levels, FP8 (E4M3) activations with FP16 accumulation, 141 MiB of weights. It is a generative neural rendering network (NVIDIA’s term): it re-renders the frame the engine already drew, generating detail from injected noise and adjusting tone, structure and skin under a style setting. Input and output are the same resolution; it is not an upscaler.

The WebGPU port at 2048x1152, NR off on the left and on on the right. Scene: Cowboy Gramps by Muhammed Ismayil, CC0.

It takes one rendered frame (a low dynamic range proxy of it, three lanes of Gaussian noise, the previous frame’s output reprojected, and five conditioning scalars) and produces four f32 channels per pixel: an RGB residual and one temporal-blend logit. docs/network.md is the graph in full.

(Excerpted from the original; full article via the source link below.)

This story hit the Hacker News front page today (239 points / 110 comments, via github.com). Our Tech Radar aggregates daily signals on AI engineering, backend architecture and DevOps — browse the related services and further reading below, or get in touch with our team.