NVIDIA Adds New AI Models, Agent Tools and Games to RTX Spark Laptops
NVIDIA announced new open-weight AI models, GPU resource controls and game support for its RTX Spark Windows laptops, now sold by six manufacturers.
Quick answer
What new AI models, agent tools and games has NVIDIA added to RTX Spark laptops?
NVIDIA announced RTX Spark laptop updates including the open-weight Nemotron 3.5 Transcribe speech model, Liquid AI's LFM3 model family, new GPU resource-sharing controls in NVIDIA App, the launch of 'Phantom Blade Zero' with DLSS 4.5, and RTX Remix 1.6 support for Minecraft, Portal with RTX and the Half-Life 2 RTX demo.
Key takeaways
- NVIDIA says RTX Spark laptops are now sold by ASUS, Dell, HP, Lenovo, Microsoft and MSI, combining a Blackwell RTX GPU and Grace CPU on one superchip with up to 128GB of unified memory and up to 1 petaflop of FP4 AI performance.
- NVIDIA introduced Nemotron 3.5 Transcribe, an open-weight speech model it says handles transcription, speaker labeling and translation across 40 languages, and says the model ranks No. 1 on Hugging Face's Open ASR leaderboard and No. 2 among open-weight models on Artificial Analysis.
- Liquid AI's LFM3 model family, spanning text, vision, audio, omni and decision models, gets day-zero support on RTX Spark, according to NVIDIA.
- NVIDIA added resource management controls to NVIDIA App that let users choose how GPU power is split between foreground apps and background AI agents.
- NVIDIA says S-GAME's 'Phantom Blade Zero' launched Oct. 28 with ray tracing and DLSS 4.5, and that RTX Remix 1.6 now brings path tracing and DLSS to Minecraft, Portal with RTX and the Half-Life 2 RTX demo.
NVIDIA announced a round of updates for its RTX Spark Windows laptops, adding new open-weight AI models, a GPU resource-management feature, and support for new games, the company said in a blog post.
RTX Spark laptops are sold by ASUS, Dell, HP, Lenovo, Microsoft and MSI, according to NVIDIA. The systems pair an NVIDIA Blackwell RTX GPU with an NVIDIA Grace CPU on a single superchip, with up to 128GB of unified memory and up to 1 petaflop of FP4 AI performance, NVIDIA says. The company says the laptops are as thin as 14mm and as light as three pounds, with all-day battery life, and come in a range of screen sizes, colors, and configurations including touchscreens, pen support and two-in-one convertibles.
New Speech Model
NVIDIA introduced Nemotron 3.5 Transcribe, which it describes as an open-weight speech model that turns recordings into transcripts, labels individual speakers, and translates audio, all through a single promptable model rather than separate tools. The company says it transcribes 40 languages, including seven Indic languages, and can answer plain-language questions about recorded audio, such as listing decisions made in a meeting. NVIDIA says the model ranks No. 1 on Hugging Face's Open ASR leaderboard and No. 2 among open-weight models on Artificial Analysis, with roughly 4.3 billion active parameters, which the company says is 80% fewer than the top-ranked open-weight model. NVIDIA says the model is available now on Hugging Face, with quantized versions coming soon for RTX Spark, DGX Spark, Jetson AGX Thor and Blackwell data center GPUs.
Liquid AI's LFM3 Models
NVIDIA says Liquid AI's LFM3 family of open models, which spans text, vision, audio, omni and decision models built on a shared backbone, now has day-zero support on RTX Spark. The company says a new 1.7-billion-parameter base model from Liquid AI outperforms the firm's previous 2.6-billion-parameter base model. According to NVIDIA, the models support local agent workflows such as triaging inboxes, managing calendars, reading documents, and transcribing speech, and can run on RTX Spark through llama.cpp and vLLM, with local fine-tuning supported.
GPU Resource Controls
NVIDIA says it added a resource management feature to NVIDIA App that lets users choose how GPU compute is divided between a foreground application and a background-running AI agent. The setting is found under System > Advanced in NVIDIA App, and NVIDIA says it requires hardware-accelerated GPU scheduling to be enabled in Windows Settings.
Games and Modding
NVIDIA says S-GAME's "Phantom Blade Zero" launched Oct. 28 on Steam and the Epic Games Store with ray tracing and NVIDIA DLSS 4.5, including Multi Frame Generation that the company says can boost frame rates up to 6x. Separately, NVIDIA says RTX Remix version 1.6, released the prior week, brings path tracing and DLSS to Minecraft Java Edition 1.21.1, and adds DLSS 5 with 3D-Guided Neural Rendering, up to 6x Multi Frame Generation, and native RTX Spark support. The company says those same upgrades, along with a new Performance Mode, now extend to Portal with RTX and the Half-Life 2 RTX demo, and that RTX Remix 1.6 includes a redesigned ComfyUI workflow for generating materials without API setup.
Source: NVIDIA Blog, "Thin, Light and Built for Everything: NVIDIA RTX Spark Windows Laptops Gain New Models, Agent Tools and Games," published Oct. 28, 2026.
Frequently asked questions
- What is NVIDIA RTX Spark?
- According to NVIDIA, RTX Spark is a laptop platform that pairs a Blackwell RTX GPU with a Grace CPU on a single superchip, offering up to 128GB of unified memory and up to 1 petaflop of FP4 AI performance.
- What does Nemotron 3.5 Transcribe do?
- NVIDIA says it is an open-weight speech model that can transcribe recordings, label speakers and translate audio across 40 languages, including seven Indic languages, and answer questions about the audio using a single promptable model.
- What is the new GPU resource management feature for?
- NVIDIA says the feature, found in NVIDIA App under System > Advanced, lets users decide whether to prioritize a foreground application or a background-running AI agent for GPU resources, and it requires hardware-accelerated GPU scheduling enabled in Windows.
- How much do RTX Spark laptops cost?
- NVIDIA says RTX Spark laptops start at $2,599, with 128GB memory configurations now available, from ASUS, Dell, HP, Lenovo, Microsoft and MSI.