Nvidia Rolls Out Software Blitz Ahead Of RTX Spark Laptop Launch

Nvidia is unleashing a wave of software updates, new AI models and gaming enhancements in the lead-up to RTX Spark laptops going on sale on October 16, with releases beginning this week and continuing into November.

The updates span local AI tools, new language models designed for on-device agents, creative application optimisations, DLSS 5 performance improvements and a string of new game integrations. Together, they form the software ecosystem that Nvidia is building around its new RTX Spark platform, which pairs a Blackwell RTX GPU with a Grace CPU in a laptop form factor.

New Nemotron Models For Local AI Agents

Nvidia is introducing two new Nemotron 3.5 Bolt models designed for efficient local AI inference.

The first, Nemotron 3.5 Bolt 74B A9B, is a mixture-of-experts model with 74 billion parameters and 9 billion active parameters. It is designed for always-on personal agents handling tasks such as coding, research, monitoring and tool use, and fits within a 24 GB memory footprint. The model is optimised for RTX Spark and will be available from October 15 with support for llama.cpp, Ollama, vLLM and LM Studio.

The second model, Nemotron 3.5 Bolt E4B, is a smaller model aimed at responsive conversational agents and interactive game characters. It arrives on October 23. Developers and gamers will be able to test the model using SkyrimNET, an AI NPC modding platform for The Elder Scrolls V: Skyrim that enables dynamic conversations and interactions with in-game characters.

Local AI Tools And Hugging Face Partnership

Nvidia and its ecosystem partners are also rolling out new tools to make running AI models and agents locally more accessible.

Hugging Face is launching Llama Windows, a free, open-source Windows application that lets users discover, download and run open AI models locally using llama.cpp without needing to use the command line. The application is optimised for RTX Spark.

Nvidia PAIR, a free, open-source tool that allows users to share AI compute across compatible PCs on a local network, has moved out of beta. The latest release adds RTX Spark and llama.cpp support, enabling users to offload AI workloads to another device while keeping inference local.

New agent integrations, MCP servers and skills are also being released that connect agents to Nvidia applications including Broadcast, the Nvidia app, RTX Remix and Omniverse. Updates across tools including Hermes, OpenClaw and Perplexity further expand what can be done with local AI on RTX hardware.

Creative Application Updates

On the creative software front, Nvidia Broadcast 2.3 is available now. The update introduces offline file processing for prerecorded content, beta RTX Video Super Resolution and Frame Generation, new Studio Voice profiles and MCP support that lets compatible AI agents control Broadcast features.

Adobe is optimising Photoshop, Premiere and Substance 3D for RTX Spark, with the company claiming up to 2X faster performance across AI, editing, colour and effects workflows in supported applications. Further features and optimisations are expected to roll out in the coming weeks.

ComfyUI, the node-based AI workflow tool, is gaining new RTX Video Frame Generation and HDR nodes, giving creators more options for enhancing AI-generated video, increasing frame rates and preparing HDR content for editing and colour grading.

Blender is also set to receive DLSS Ray Reconstruction support in version 5.3, expected in November, which is intended to improve viewport interactivity.

RTX Spark Launches With Broad Game Support

When RTX Spark laptops go on sale on October 16, they will support hundreds of PC games out of the box, including CONTROL Resonant, Cyberpunk 2077, Fortnite, Gears of War: E-Day, Genshin Impact, Hogwarts Legacy, Honkai: Star Rail, Marvel Rivals, NARAKA: BLADEPOINT, NBA 2K27 and World of Warcraft.

The platform supports the full suite of RTX gaming technologies, including DLSS 5 with 3D-Guided Neural Rendering, ray tracing, Nvidia Reflex and G-SYNC. Nvidia has also confirmed compatibility with major multiplayer anti-cheat systems.

DLSS 5 Gets A Faster AI Model

Nvidia is pushing out a new, faster DLSS 5 model that the company says is 2X faster than the previous version, with overall performance now 10X faster than the initial model introduced in March.

The updated DLSS 5 with 3D-Guided Neural Rendering model arrives on October 16 alongside NBA 2K27 Season 2, where it is expected to boost frame rates by up to 1.4X on supported GeForce RTX 50 Series GPUs, RTX Spark laptops and GeForce NOW.

Hogwarts Legacy will add DLSS 5 support on November 5, while Where Winds Meet is expected to receive a DLSS 5 update in late November.

RTX Remix is also getting an update on October 23 that adds DLSS 5 support for modders, along with streamlined AI-powered texture generation workflows through ComfyUI integration and batch processing.

New DLSS Game Integrations

Beyond DLSS 5, Nvidia is continuing to expand the library of DLSS-accelerated games across its older DLSS versions.

Call of Duty: Modern Warfare 4 Campaign Early Access begins on October 16 with DLSS 4.5 support. Crossout Remastered’s DLSS 4.5 update arrives on October 22. Among The Trolls launches on October 14 with DLSS and ray tracing.

Several titles are already available with DLSS support, including Gear.Club Unlimited 3 with DLSS and path tracing, Permafrost with DLSS 4.5 and ray tracing, Pax Autocratica with DLSS 4.5 and ARC Raiders’ Frozen Trail update with DLSS.

Nvidia App Gets A New Performance Mode

A new Nvidia app beta update introduces Preset C, a faster performance mode for DLSS Frame Generation, Multi Frame Generation and Dynamic Multi Frame Generation.

The new mode prioritises higher frame rates using a lightweight AI model and is available across GeForce RTX 40 Series, RTX 50 Series and RTX Spark laptops. The beta is available now, with the new preset applicable to hundreds of existing games and applications through the Nvidia app.

Last Updated on October 14, 2026 by Nick Ross