THE APEX TIMES
New compact Windows PCs from ASUS, Dell and MSI bring NVIDIA’s Grace-Blackwell platform, up to 128GB of unified memory and a software stack aimed at running AI agents and generative workflows locally.
NVIDIA says it is expanding the RTX Spark hardware lineup with new compact desktop PCs that target developers and creators who want to run AI agents and generative workloads on their own machines. The company’s announcement, published as the systems become available for purchase, frames the desktop move as a step toward “always on” local work, where tasks such as image generation, video pipelines and scheduled model fine-tuning do not have to wait for a cloud connection.
The RTX Spark compact desktops are designed to pair a dedicated CPU and GPU setup with substantial memory capacity. NVIDIA says the systems include a 20-core NVIDIA Grace CPU, an NVIDIA Blackwell RTX GPU with 6,144 CUDA cores, and up to 128GB of unified memory. NVIDIA also highlights a high-speed 10-gigabit Ethernet connection to speed transfers of large datasets and model files between the desktop and networked storage. NVIDIA characterizes performance as up to 1 petaflop of AI throughput.
A core part of NVIDIA’s pitch is that local execution can keep code and files on the PC instead of routing per-token workloads to a cloud API. In practical terms, NVIDIA is positioning the RTX Spark desktop as a dedicated workstation for workflows where creators and developers want to continue working while agents run in the background or process longer renders after an active session ends.
NVIDIA’s software story focuses on what it calls an expanding ecosystem for RTX Spark laptops and compact desktops. The company points to native integration in Windows environments and to AI tooling that can coordinate local and cloud model use. It also cites support for Windows Subsystem for Linux to give developers Linux tooling under Windows, along with CUDA support for building and running local AI workflows.
On the agent side, NVIDIA points to Hermes Agent and NVIDIA NemoClaw blueprints as ways to set up agent behaviors and then fine-tune supported models using Unsloth. For routing tasks, NVIDIA also describes NVIDIA Switchyard, which it says is now available in Hermes. The company describes Switchyard as a system that can automatically route agent work between on-device and cloud models, aiming to run everyday tasks locally while using cloud models for more advanced reasoning, with the goal of managing cloud costs across creative workflows.
For creative pipelines, NVIDIA highlights how RTX Spark connects familiar tools with local generation software. One example described by NVIDIA is the use of Hermes Agent to connect Photoshop and ComfyUI using natural-language instructions, allowing creators to move from sketches to rendered images and animations. NVIDIA also describes an RTX video path built into ComfyUI, including RTX Video Frame Generation and RTX Video HDR, which it says can extend generated clips to 60 frames per second and convert them to high dynamic range within a single ComfyUI workflow.
The hardware and software story is also tied to tighter integration with Adobe products. NVIDIA says native Windows on Arm applications and RTX acceleration are expanding across Adobe’s creative suite for RTX Spark systems. It cites TensorRT-RTX acceleration in Adobe Premiere for features such as Enhance Speech, Object Mask, Scene Edit Detection and Auto Reframe. NVIDIA also says the update improves transitions and effects plug-in performance and provides expanded support for professional video formats. For photography and design, NVIDIA says Adobe Photoshop adds TensorRT-RTX acceleration for the Remove Tool and adds Arm-native Camera Raw support, while Adobe Substance 3D Painter is optimized for RTX Spark and Adobe Illustrator and InDesign offer native Windows on Arm versions optimized for RTX Spark.
Outside Adobe’s stack, NVIDIA points to Blender updates aimed at interactive ray-traced previews. It says Blender 5.3 integrates DLSS 4.5 Ray Reconstruction into the Cycles viewport, describing the feature as delivering near-final-quality lighting, reflections and materials in real time rather than a noisy preview. NVIDIA adds that DLSS 4.5 Ray Reconstruction support requires Blender 5.3 and the latest NVIDIA Game Ready Driver or NVIDIA Studio Driver, and it points users to its NVIDIA App for driver notifications and unified control.
NVIDIA’s broader Windows platform messaging frames the desktops as one part of a wider push to run more AI workloads in enterprise and creator environments. The company also references updates around Windows Server capabilities announced at Microsoft Ignite and mentions developer options for running models in Windows ML using a backend for GGUF models from Hugging Face. It additionally describes experimental work for game developers through a DxCGC backend that compiles and runs machine learning models via DirectX 12, and it points to APIs that integrate platform-provided AI capabilities alongside GPU support for additional functions.
What NVIDIA does not detail in the announcement are the specific retail configurations, pricing, and detailed availability windows for each OEM. The company also does not provide benchmark breakdowns comparing RTX Spark desktops to alternative workstations, nor does it spell out which exact models are supported for agent fine-tuning and local routing. For buyers and developers, the next question is whether the advertised local-and-cloud task routing and creative agent workflows translate consistently across real projects, and how quickly the wider Windows-on-Arm software ecosystem continues to expand for RTX Spark.
Investors moved quickly after early outlines connected United Launch Alliance’s work to the BE-4 engine family, which Blue Origin’s New Glenn is expected to use for a planned 2027 flight carrying AST SpaceMobile’s BlueBird satellites.
A report tied to Anthropic’s debut highlights a cash position large enough to dwarf the combined figures of Amazon and NVIDIA, while also pointing to a new exchange-traded fund built around companies described as “ANT-linked.”
A Yahoo Finance report highlights how Berkshire Hathaway’s roughly $10 billion investment in Alphabet appears to be losing value, with Bill Ackman commenting that the timing and pricing were unexpected even as he noted the stake was bought at a perceived discount.
The company said it has reshaped leadership around national security and export-control oversight after Huang’s exit, emphasizing governance changes ahead of its next quarter and the continuing geopolitical scrutiny on advanced chips.
Anthropic priced its IPO at $83 per share ahead of its debut, while Nvidia heads toward a seventh straight weekly decline amid fresh US pressure on AI chip exports.
A report citing Cathie Wood’s ARK funds says they sold about $84.8 million of Nvidia shares before the company’s earnings period and used the cash to buy holdings tied to crypto markets.
Reports tied to Nvidia’s former chief executive, including allegations of multiple affairs involving a suspected Chinese intelligence-linked figure, are now being read alongside a broader, politically sensitive backdrop for advanced chip sales across the United States and China.
A celebrity investor publicly criticized NVIDIA’s former CEO Jensen Huang, then appeared to reverse course several days later, echoing a message that Nvidia employees “know better than the board,” according to a Yahoo Finance report.
A reporting update says the major banks and asset managers that earlier this year promised as much as $500 billion in financing support for Nvidia customers have not yet moved capital into that program.
The sale, disclosed in a regulatory filing, trims Lisa Su’s direct stake in Advanced Micro Devices while the company’s shares have climbed sharply this year.
Comments
Community guidelines. Comments may be removed for violations.
Loading comments...