NVIDIA introduced 2nd Generation Ray Reconstruction as a component of DLSS 4.5. The addition fills a gap in the DLSS 4.5 feature set after other parts of the suite had already moved to 2nd Gen transformer models trained on larger sets of data.
When Ray Reconstruction debuted in 2023 as part of the DLSS 3.5 update, it addressed an image quality issue where ray traced shadows and reflections appeared inaccurate when used with DLSS 2 super resolution. In 2025, NVIDIA updated the older convoluted neural networks (CNN) based Artificial Intelligence model with a newer transformer based model. Later that year, when NVIDIA introduced the DLSS 4.5 feature-set with 2nd Gen transformer models trained on larger sets of data, Ray Reconstruction didn’t get updated.
The company has now completed the DLSS 4.5 feature-set with the new Ray Reconstruction feature. Besides a 2nd Gen transformer model trained on a larger volume of data, DLSS 4.5 RR introduces a more efficient Artificial Intelligence denoiser, and greater spatial awareness for lighting accuracy.
