Skip to main content
Apex Neural Rendering Challenges NVIDIA DLSS With Up To 2.4X Faster Frame Generation, 62% Lower VRAM Usage, And Broad GPU Support - Wccftech

Apex Neural Rendering Challenges NVIDIA DLSS With Up To 2.4X Faster Frame Generation, 62% Lower VRAM Usage, And Broad GPU Support - Wccftech

NVIDIA GPUs
12 lis 2026, 12:25Hardware News
Oryginał
2 min read

Apex Compute has introduced Apex Neural Rendering (ANR), a cross-platform neural rendering suite that reportedly delivers up to 2.4X faster frame generation and 62% lower VRAM usage than NVIDIA's DLSS technology.

NVIDIA DLSS may have a promising new competitor. Apex Compute has introduced Apex Neural Rendering (ANR), a cross-platform neural rendering suite that reportedly delivers up to 2.4X faster frame generation, superior image quality, and substantially lower GPU memory consumption than NVIDIA's technology. Better still, gamers can already try it for themselves — or at least its (Multi) Frame Generation component, which is currently available in beta.

ANR consists of three components: Frame Generation (FG), Super Resolution (SR), and Ray Reconstruction (RR). Frame Generation is currently available in beta, while its upscaling and ray-reconstruction models remain in development.

Broad Compatibility

One of ANR's most interesting features is its unusually broad compatibility, encompassing considerably older hardware that lacks official NVIDIA DLSS (Multi) Frame Generation support. The technology works across NVIDIA GeForce GTX 16 and RTX 20/30/40/50 Series, AMD Radeon RX 5000/6000/7000/9000 Series and Steam Deck, and Intel Arc A-Series and B-Series.

The beta supports Windows 10/11 alongside Linux and SteamOS through Proton. Apple silicon is also supported through a separate native Metal test kit for macOS 26 and newer.

Performance Claims

Performance looks promising, at least according to Apex Compute's internal benchmarks. On a GeForce RTX 5090 at 3840×2160, ANR FG requires 1.01 milliseconds per generated frame in 2X mode, compared to 1.26 ms for NVIDIA DLSS Frame Generation. In 4X mode, the figures drop to 0.61 ms and 0.82 ms, respectively.

GPU memory consumption is another claimed advantage, with ANR requiring just 214 MB versus NVIDIA's 562 MB at 4K, representing a reduction of approximately 62%. Its neural network weights are also remarkably compact, occupying merely 0.17 MB.

Although Apex advertises up to 2.4X faster frame generation in certain tests, its published RTX 5090 measurements suggest more modest processing-time improvements. These results concern the cost of generating additional frames rather than equivalent increases in actual gaming FPS.

Image Quality Advantages

Image quality is another major selling point. Apex's interactive comparisons demonstrate improvements in challenging scenarios involving particles, moving objects, water, vegetation, and disocclusion artifacts. The official ANR website features numerous side-by-side frame-generation comparisons against NVIDIA DLSS (M)FG, highlighting these improvements.

Z.ai
Article written by AI
Z.ai GLM-4.5-FlashFree
Free AI articles150 of 150
No limit after signing in with Telegram