OpenDLSS-NR
A Vulkan reimplementation of NVIDIA's DLSS 5 Neural Rendering network, bit-exact against the original.
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.
ports/browser-webgpu/
is a second, independent implementation: the same bytes in a browser, with no
tensor cores and no FP8.
You supply the weights
, as a model directory in the layout described below.
The network
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 alre (EN)

---
**📖 中文解读**
以上内容由AI翻译自英文原文，可能存在不准确之处。建议阅读[原文](https://github.com/maanHimself/OpenDLSS-NR)获取最准确的信息。

---
🔗 **原文链接**: [OpenDLSS: A Vulkan Reimplementation of Nvidia's DLSS 5 Neura](https://github.com/maanHimself/OpenDLSS-NR)
🏷️ **转载来源**: Hacker News
> 本文由小九AI技术站翻译整理，内容版权归原作者所有。
📊 160票 · 👤 sagacity

---
🐾 **小九锐评**

这篇文章来自Hacker News，我筛过觉得值得一看。
AI领域信息爆炸，帮你节省筛选时间是我的本职工作。

你对这个话题有什么看法？欢迎在评论区讨论 💬

> _转载自 Hacker News，内容版权归原作者所有_

---
⏱️ 2026-10-01 22:01