AI image generation is now 7x faster on Windows PCs with NVIDIA GPUs

Generative AI, one of the most significant trends in personal computing history, is now up to 4x faster on Windows PCs with NVIDIA’s open-source library, TensorRT-LLM for Windows. This advancement is set to revolutionize gaming, creativity, video, productivity, development, and more. NVIDIA’s GPUs come with dedicated AI processors known as Tensor Cores.

TensorRT-LLM will soon be available to download from the NVIDIA Developer website. TensorRT-optimized open-source models are available at ngc.nvidia.com and GitHub.com/NVIDIA.

TensorRT accelerated AI models through layer fusion, precision calibration, kernel auto-tuning and other capabilities that significantly boost inference efficiency and speed. With TensorRT, you can now expect double the speed of Stable Diffusion AI image generation. When compared to Apple M2 Ultra macs, GeForce RTX 4090 Windows PCs runs Stable Diffusion AI image generation 7 times faster. The extension is available for download today.

In addition, RTX VSR improves the quality of streamed video content from YouTube, Twitch, Prime Video, Disney+ and others by reducing or eliminating artifacts caused by video compression. It also sharpens edges and details. According to NVIDIA, the RTX VSR version 1.5 improves visual quality with updated models, de-artifacts content played in its native resolution and adds support for RTX GPUs based on the NVIDIA Turing architecture — both professional RTX and GeForce RTX 20 Series GPUs.

RTX VSR 1.5 can be downloaded with the latest Game Ready Driver. It will be available in the upcoming NVIDIA Studio Driver early next month.