Nvidia RTX Spark: 1-Petaflop Superchip Brings Local AI Agents to Windows PCs
Nvidia today announced RTX Spark, a new superchip that brings 1 petaflop of AI compute to Windows PCs, purpose-built for running personal AI agents locally. The chip combines a Blackwell RTX GPU (6,144 CUDA cores, fifth-gen Tensor Cores with FP4), a 20-core Grace CPU, and up to 128GB of unified memory, connected via NVLink-C2C. MediaTek contributed to the custom Arm-based CPU design for power efficiency.
Key Specs and Capabilities
- AI performance: 1 petaflop (FP4)
- GPU: Blackwell RTX with 6,144 CUDA cores
- CPU: 20-core NVIDIA Grace (Arm), co-designed with MediaTek
- Memory: up to 128GB unified memory
- Software stack: CUDA, RTX, DLSS, FP4, TensorRT, OptiX, Reflex, G-SYNC
RTX Spark can run 120B-parameter LLMs with up to 1 million tokens context locally, render 90GB+ 3D scenes, edit 12K 4:2:2 video, generate 4K AI video, and play AAA games at 1440p 100+ fps.
Windows-Native Agent Security
Nvidia and Microsoft are collaborating on new Windows security primitives and the Nvidia OpenShell runtime to enable secure on-device agents. The security layer provides identity, containment, policy, and end-to-end security. OpenShell adds user-defined policies for agent capabilities, intelligent query routing to local vs. cloud models, and PII masking in cloud-bound queries.
Agent frameworks including Hermes Agent and OpenClaw are building Windows apps on this stack, enabling cross-app workflows, file search, image/video generation, and code plugin creation.
Availability
RTX Spark-powered slim laptops (all-day battery, premium displays) and compact desktops will ship this fall from ASUS, Dell, HP, Lenovo, Microsoft Surface, and MSI, with Acer and GIGABYTE models to follow.
📖 Read the full source: HN AI Agents
👀 See Also
Hollywood Writers Shift to AI Training: First-Person Account of Data Annotation Work
A Hollywood showrunner describes transitioning to AI training work at $52/hour after the 2023 strike, annotating conversations, images, and videos for companies like Mercor and Outlier.
Trading Strategy Benchmark: Cheaper AI Models Outperform Claude Opus 4.6
A benchmark tested 10 LLMs on developing trading strategies, with cheaper models like Minimax 2.5 and Gemini 3.1 outperforming Claude Opus 4.6 despite its 10x higher cost. The experiment was run three times with consistent results.
Anthropic files lawsuit to prevent Pentagon blacklisting over AI restrictions
Anthropic has filed a lawsuit seeking to block the Pentagon from blacklisting the company over restrictions on AI use, according to a Reuters report shared on Hacker News.
Anthropic Clarifies Claude CLI Usage Policy for OpenClaw Integration
Anthropic has confirmed that OpenClaw-style Claude CLI usage is permitted again, allowing developers to reuse existing Claude CLI logins directly. The documentation details both API key and CLI authentication methods, along with configuration options for Claude 4.6 models, fast mode, and prompt caching.