GoKawiilModders port leaked DLSS 5 Neural Rendering to RTX 4000 GPUs
A day after DLSS 5's nvngx_dlssnr.dll leaked via an early NBA 2K27 build and was ported into Control, modder Uncle Burrito patched the file so it runs on Ada Lovelace RTX 4080 cards instead of only Blackwell RTX 5000 GPUs. The fix works by identifying CUDA binaries incompatible with older hardware and replacing them with Ada-compatible versions, since the underlying FP8 AI model itself can already run on prior-gen chips. Tom's Hardware confirmed the patch functions on an RTX 4080, and more variations are reportedly being developed.
GoKawiil's interpretation of the reporting above, not reported fact.
This shows Nvidia's newest Neural Rendering feature isn't fundamentally locked to Blackwell hardware, undercutting any hardware-exclusivity marketing tied to DLSS 5 before its official release. It also highlights how quickly community modders can reverse-engineer leaked Nvidia tools, potentially forcing Nvidia to rethink how it stages feature rollouts across GPU generations.
- Leaked DLSS 5 DLL from NBA 2K27 has been reverse-engineered to run on RTX 4000 series cards.
- The block on older GPUs was due to incompatible CUDA binaries, not the AI model itself.
- Community modder Uncle Burrito patched the binaries, with Tom's Hardware confirming it works on an RTX 4080.
NVIDIA GeForce RTX 5070 Ti Graphics Card — If you'd rather run DLSS 5 neural rendering natively instead of relying on community DLL patches, a Blackwell-generation card like the RTX 5070 Ti is the straightforward path. It's the hardware the new DLSS binaries were built to target, so you skip the ReShade/RenoDX workarounds entirely.
See NVIDIA GeForce RTX 5070 Ti Graphics Card on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: tomshardware.com — Sayem Ahmed, 2026-08-28
Published there as: “DLSS 5 has already been ported to work on RTX 4000 Series graphics cards — incompatible CUDA instructions get patched to work on previous-gen hardware”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.