NVIDIA has expanded its developer toolkit with new software updates for RTX hardware and artificial intelligence workflows. The company announced TensorRT for RTX, a plugin for Unreal Engine, alongside support for the ComfyUI platform and the DLSS 4.5 SDK. These releases aim to provide developers with improved tools for integrating RTX and AI features into modern applications and games.
The TensorRT for RTX plugin for Unreal Engine's Neural Network Engine delivers a 1.5x performance improvement compared to DirectML-based approaches. Developers can deploy AI models directly within applications for better efficiency. The software supports all RTX GPUs ranging from the Turing architecture to the Blackwell architecture. ComfyUI is now available on Windows and Linux systems for any NVIDIA RTX GPU with 16GB or more of VRAM.
NVIDIA also introduced the DLSS 4.5 SDK, which supports Dynamic Multi-Frame Generation and Multi-Frame Generation 6X. The company highlighted its investment in Neural Rendering technology, claiming a 1,000,000x performance improvement over current generation GPUs. New algorithms allow path tracing to run three times faster. Neural Texture Compression technology reduces VRAM usage by 85% while improving image quality.
NVIDIA is betting heavily on the RTX GPU and AI PC ecosystem. The vendor is providing developers with new and improved toolsets to accelerate content creation. The announcement was made during the NVIDIA GTC 2026 developer institute session. The company emphasized that these tools are designed to help developers integrate advanced graphical capabilities into their software.
Discussion
0 comments
Log in to join the thread with a thoughtful take, question, or correction.