Nvidia launches RTX Spark for local AI agents on PCs - ForkLog: cryptocurrencies, AI, singularity, the future

Nvidia# Nvidia unveils RTX Spark for local AI agents on PCs

Nvidia introduced RTX Spark — a chip for Windows PCs focused on running AI agents and large language models locally. The project has been developed together with Microsoft for about three years

Create, build, and play: https://t.co/jYkBa10Mq8

— NVIDIA (@nvidia) June 1, 2026

The company positions RTX Spark as the foundation for devices for personal AI systems that operate directly on the computer without necessarily relying on the cloud.

The claimed performance is up to 1 petaflop of computational power and up to 128 GB of unified memory. Nvidia expects users to be able to run agent-based and generative models directly on their PCs.

The chip is built on the Blackwell RTX GPU with 6,144 CUDA cores and a 20-core Grace processor. MediaTek participated in the creation of the CPU. The platform is designed for Windows 11 and scenarios involving AI agents, content creation, and gaming.

Microsoft joins the agent-based Windows

Nvidia and Microsoft announced joint work on infrastructure for agent systems within Windows.

The release mentions new security mechanisms and the NVIDIA OpenShell platform for launching agents on user devices. This refers to preparing Windows for such scenarios, not a full transition of the OS to an agent-based model.

Microsoft is already developing this direction through Copilot and enterprise products, where assistants automate workflows.

Nvidia CEO Jensen Huang called this a shift from “PC as a tool” to “PC as an assistant.” Essentially, this is about launching a new hardware platform for local AI.

Adobe optimizes Photoshop and Premiere

Adobe is adapting Photoshop and Premiere for RTX Spark. Nvidia claimed a twofold increase in AI and graphics performance on the new platform. There are no independent confirmations of these figures.

In May, Adobe added an on-device model to Photoshop 27.7 for the Remove tool — objects can be removed without sending data to the cloud. Users can choose between local and cloud processing.

For Premiere, the company introduced AI features, including Object Mask, but has not announced a full transition of the video editor to local AI.

Release — Fall 2026

The first PCs based on RTX Spark will be released in fall 2026. Partners include ASUS, Dell, HP, Lenovo, Microsoft Surface, and MSI. Later, Acer and GIGABYTE will join.

Nvidia is increasing its focus on the AI-PC segment, where AI agents, data analysis, and content generation are performed locally — without constant dependence on the cloud.

Recall that in March, the company announced the launch of infrastructure for creating data centers in space.

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pinned