DLSS 5 Neural Rendering Now Works on RDNA2-4 GPUs via Open-Source Mod

A community developer has released a modification that allows AMD Radeon graphics cards to use Nvidia’s DLSS 5 features, including Neural Rendering. This project, known as DLSS-NR on AMD, was first released by Daniel Blanco to address the lack of AMD equivalents for neural rendering. The latest update expands this capability significantly. According to Tom’s Hardware, the 0.8.0 release makes the entire suite of DLSS features available to users with compatible hardware. This includes Super Resolution, Frame Generation, Ray Regeneration, and the newest Neural Rendering model.

The risk

The primary concern for users adopting this mod is potential instability or inconsistent visual output. Daniel Blanco noted that while the image quality closely matches native Nvidia performance, some effects may not function at one hundred percent capacity. The mod operates by inserting itself into the upscaling pipeline of games that support FSR, DLSS, or XeSS. This interception method means the tool relies on the game’s existing compatibility with those technologies. Any conflict between the mod’s injection process and the game engine could lead to crashes or graphical artifacts.

Who is exposed

The mod currently supports three generations of AMD Radeon GPUs: the RX 9000, 7000, and 6000 series. These correspond to the RDNA4, RDNA3, and RDNA2 architectures, respectively. The tool functions in any DirectX 12 game that utilizes FSR, DLSS, or XeSS upscaling technologies. Users must install the NeuralHub mod loader, which provides a user interface for configuring the mod within detected games. The loader allows players to enable FSR, DLSS, or XeSS in the game settings, after which the mod injects the DLSS features into the rendering pipeline.

Cost and compliance

This modification is a solo project, not an official product from AMD or Nvidia. There is no licensing fee or commercial cost associated with the DLSS-NR mod. The developer, Daniel Blanco, is making improvements every day, planning to match the performance of Nvidia’s 5070 Ti card. Testing on an RX 9070 XT showed playable frame rates between 90 and 100 FPS at 1080p resolution. The update notes specify the inclusion of FSR 4.1.1 upscaling to provide a consistent version for the mod to work with.

What should you do?

Users interested in this mod should verify their game’s compatibility with DirectX 12 and FSR, DLSS, or XeSS before installation. It is advisable to create system restore points or backup game saves prior to using the NeuralHub loader, as the mod intercepts core rendering processes. Monitor for graphical anomalies or crashes during initial use, as the developer notes some effects may not work perfectly.

The background in plain language

Graphics upscaling technologies render games at lower resolutions and use algorithms to enhance image quality to higher resolutions. This improves performance on less powerful hardware. Neural rendering uses artificial intelligence models to reconstruct images, offering higher quality than traditional methods. Mods like this intercept

What to do and how to stay safe

• Full disk encryption protects data at rest, but does not secure data while it is actively being processed in memory.
• Hardware-backed key storage prevents private keys from ever leaving the secure element, mitigating remote extraction risks.
• Disabling legacy wireless protocols like Bluetooth Classic reduces the attack surface for proximity-based exploits.

Mobile security requires a defense-in-depth approach that combines hardware features, strict permission management, and secure communication protocols. Start by enabling full disk encryption and disabling unnecessary wireless protocols to immediately reduce your attack surface.

Step-by-step guide: Mobile Security Architecture: 8 Practices for Device Integrity

General security guidance from the Vector Tribune newsroom. It is not confirmed advice from the organisations named in this story.