Chips & Hardware

First RTX Spark measurements: half the time of a MacBook Pro on AI agent tasks, but not ahead on raw output from large models. Who the chip is worth it for.

Microsoft unveils Maia 300: its primary AI inference system from 2027, with a projected 40 percent more tokens per megawatt. What the chip does and what users gain.

Nvidia brings RTX Spark to Windows laptops, and Microsoft gives agents their own security containers. What is new, what it costs, and what is missing.

Nvidia announces a 64 GB DGX Spark starting at $4,999. What memory, local models, and clustering mean for developers and smaller teams.

CoreWeave opens Vera Rubin to its first customers. Cognition reports higher throughput for Devin. What the figures measure and who should run their own comparison.

OpenAI and Synopsys are building GPT-Synopsys, an AI model for chip design tools. What it is meant to do, how the business model works and where its limits lie.

Meta unveils VR glasses weighing about 100 grams, with micro-OLED displays and a compute puck, for $1,299. What they can do, what early testers say, and the limits.

Qualcomm’s new Snapdragon chips are built to run 30-billion-parameter AI models locally. What that enables, where the limits are, and why memory matters.

Apple is reportedly building an enterprise server with its own M8 Ultra chips for AI inference — fueled by AI firms already buying up Mac hardware in bulk.

A new data center in Inner Mongolia shows how China is weaning itself off Nvidia for AI inference, even as US export controls fuel the shift.