Nvidia announced dlss 4.5 at the ces trade show, introducing dynamic multi frame generation, a new 6X multi frame generation mode and a second-generation transformer model for dlss super resolution so gamers can experience current titles with enhanced performance and visuals. Over 250 games and apps now support Nvidia dlss 4 technology, including upcoming releases such as 007 First Light, Phantom Blade Zero, Pragmata and Resident Evil Requiem at launch. The company also highlighted new capabilities across its ecosystem, from rtx remix logic for dynamic mod-driven graphics to Nvidia ace technology that uses artificial intelligence to create more responsive advisors and teammates in titles like Total War: Pharaoh and Pubg: Battlegrounds.
Dlss 4.5 can generate up to five additional frames per traditionally rendered frame, dynamically boosting performance and enabling 240+ frames-per-second gaming with path tracing using Geforce rtx 50 series gpus, and dynamic multi frame generation and 6X multi-frame generation are expected in spring of this year. The second-generation transformer model for dlss super resolution is designed to deliver state-of-the-art image quality to over 400 games and apps for all Geforce rtx gpus, and is already available to try via the Nvidia app. Since dlss 4 with multi frame generation launched at ces last year with 75 games and apps supported, the roster has grown to more than 250, with new and upgraded titles like Active Matter, Defect and Screamer adding the latest Geforce rtx features.
Nvidia is also rolling out next-generation g sync pulsar gaming monitors that use variable frequency backlight strobing to deliver effective motion clarity of over 1,000Hz, as well as g sync ambient adaptive technology that relies on a built-in light sensor to automatically tune color temperature and brightness. For modders, rtx remix logic will arrive later this month via the Nvidia app, offering 900+ configurable settings so visuals across 165+ classic games can react dynamically to in-game events without engine source access. Nvidia ace is being expanded to turn conversational npcs into autonomous characters that use artificial intelligence to perceive, plan and act, with creative assembly testing a dynamic advisor in Total War: Pharaoh and Krafton giving the Pubg ally teammate long-term memory for more contextual responses during a limited-time test in the first half of this year.
On the cloud side, Geforce now is adding updates that deliver Geforce rtx 5080-class performance to more devices, including new native apps for linux systems and Amazon fire tv sticks, plus support for hands-on throttle-and-stick flight control peripherals and streamlined single sign-on for Gaijin titles. Nvidia is also positioning rtx artificial intelligence pcs as a local hub for generative workloads, citing up to 3x performance and 60% reduction in vram for video and image generative artificial intelligence via Pytorch-cuda optimizations and native nvfp4/fp8 precision support in Comfyui, rtx video super resolution integration in Comfyui to accelerate 4K video generations, and a blueprint for generating 4K artificial intelligence video from a 3d scene in Blender. Additional improvements include Nvidia nvfp8 optimizations for Lightricks’ Ltx-2 audio-video generation model, up to 35% faster inference performance for slms via Ollama and llama.cpp and rtx acceleration for Nexa.ai’s Hyperlink video search, all aimed at enabling advanced video, image and language artificial intelligence workflows to run locally with greater privacy and lower latency.
