Back
NVIDIA RTX Spark PC Platform Prepares for Launch With Ecosystem Software Updates
SiTech AI Team3 min read

NVIDIA RTX Spark PC Platform Prepares for Launch With Ecosystem Software Updates

Ahead of the October 16 launch of RTX Spark laptops, NVIDIA and partners rolled out agent app updates, new models, creative app optimizations and developer tools for the local AI supercomputer platform.

Platform Launch Approaches With Ecosystem Updates

NVIDIA RTX Spark combines the capabilities of a gaming PC, a creative studio and a personal AI supercomputer into a single superchip with up to 128GB of unified memory and 1 petaflop of AI compute. Announced with Microsoft, RTX Spark laptops become available on Friday, Oct. 16, and NVIDIA and its ecosystem partners are rolling out software features and performance optimizations ahead of the launch.

Local AI Agents Get New Models and Tools

Updates to agent apps including GitHub Copilot, Hermes, OpenClaw, Perplexity, Pi and WorkBuddy streamline setup on RTX Spark and add security through NVIDIA OpenShell and Microsoft Execution Containers (MXC), which delivers operating system level security. NVIDIA Switchyard, an open source model router library, enables agents to route queries between local and cloud models. The Nemotron 3.5 Bolt 74B model, purpose built for agentic AI, releases on Thursday, Oct. 15. It is a 74B parameter, 9B active parameter mixture of experts model packed into 24GB using 2-bit quantization, and will be available through llama.cpp, vLLM, Ollama, Unsloth and LM Studio. NVIDIA PAIR, a personal AI router that distributes workloads across PCs on a local network, is moving out of beta with llama.cpp support, direct Hugging Face model downloads and RTX Spark support.

Creative Apps and Developer Tools

Adobe is bringing up to 2x performance improvements to Photoshop, Premiere and Substance 3D Painter through GPU accelerated compositing and faster playback, rendering and effects, with beta builds available now. NVIDIA Broadcast 2.3 adds Video Super Resolution, Frame Generation, an offline file processing mode and MCP controls for local agents. ComfyUI gains RTX Video Frame Generation and HDR nodes, while Blender 5.3 adds RTX Spark compatibility and DLSS Ray Reconstruction arrives in November. On the developer side, PyTorch, Python 3.15, CUDA and over 2,000 Conda packages now have native Windows on Arm builds, with 97% of the top 15,000 Python wheels ready. Llama.cpp adds support for decision models and releases Llama Windows, a Windows native app for downloading and running models locally.

Gaming: DLSS 5 and Hundreds of Titles

Hundreds of top games including Cyberpunk 2077, Fortnite, Genshin Impact and Hogwarts Legacy are supported at launch, along with anti-cheat support for popular multiplayer titles. DLSS 5 with 3D-Guided Neural Rendering arrives on launch day, and a 2x faster DLSS 5 AI model debuts in NBA 2K27 on Oct. 16, also coming to Hogwarts Legacy on Nov. 5 and Where Winds Meet in late November. The third generation Nemotron 3.5 Bolt E4B 4GB small language model for NVIDIA ACE game characters releases on Friday, Oct. 23, alongside an RTX Remix update with DLSS 5 support and ComfyUI Version 2 for texture generation. SkyrimNET, an AI NPC modding platform, now supports RTX Spark with local and cloud inference.

Sources: Nvidia Blog

SSiTech

SiTech — AI-powered web development

We build fast, modern websites and bring AI into real business workflows. Have a project or a question? We'd love to help.