The First CPU Built for AI Agents

Nvidia announced on August 24, 2026 that SpaceXAI, the AI company linked to SpaceX and xAI, will adopt Vera, described by Nvidia as the market's first CPU designed specifically to accelerate artificial intelligence agents. The announcement was released officially by Nvidia and via GlobeNewswire, confirmed by outlets including StorageReview, WCCFTech, Seeking Alpha, StockTitan and Investing.com.

Up to 1.8 Times Faster Than x86 CPUs

According to Nvidia, Vera delivers up to 1.8 times the performance of traditional x86 CPUs across agentic AI, reinforcement learning and data processing workloads. The chip packs 88 Nvidia designed Olympus cores, Nvidia Spatial Multithreading technology, and high bandwidth LPDDR5X memory capable of delivering up to 1.2 terabytes per second. According to the company, agentic AI applications increasingly rely on CPUs to orchestrate tools, execute code, process data and run simulations between calls to the AI model, a role Vera was designed to fill.

Grok Also Gets More Firepower With Vera Rubin

Beyond adopting the new CPU, SpaceXAI is expanding the AI infrastructure behind Grok with the Nvidia Vera Rubin platform, as the company scales toward gigawatts of computing capacity. The move expands the processing capacity available to train and run future versions of xAI's AI model.

Starmind: Artificial Intelligence Headed to Orbit

The most ambitious part of the announcement is extending this infrastructure beyond Earth: SpaceXAI plans for Starmind, its first generation AI satellite, to be based on an Nvidia Vera Rubin NVL72 system optimized specifically for orbital operation. The initiative signals an intent to bring cutting edge AI processing capacity directly to satellites, not just to ground based data centers.

Why It Matters for Brazilian Agencies and SMBs

Vera's launch reinforces a relevant trend for anyone building agentic AI products: increasingly, an autonomous agent's performance depends not only on the language model itself, but also on the CPU responsible for orchestrating tools, executing code and processing data between one model call and the next. For Brazilian agencies running customer service and automation agents at scale, the emergence of hardware dedicated to this orchestration layer is likely to translate, over time, into faster and cheaper cloud plans for running complex agents, which is worth tracking when negotiating infrastructure contracts with providers that already adopt or plan to adopt CPUs like Vera.