AI & Tech

Large hybrid MoE models like Nemotron-3-Super are accurate but expensive to serve. Their active parameters, KV cache, and Mamba state cap how many users a node can hold at a given per-user token rate. NVIDIA AI team has released Nemotron-Labs-3-Puzzle-75B-A9B, a compressed variant of Nemotron-3-Super. The parent model has 120.7B total and 12.8B active parameters….

Manna Aero, the Ireland-based autonomous drone delivery startup, has been a smaller player in the United States. Founder and CEO Bobby Healy told TechCrunch that’s about to change. The startup, fueled by the $50 million in venture capital it raised in April, said Wednesday that it’s setting up a U.S. operations and manufacturing center in…

The race to produce more chips is on, and Europe is in. ASML, the Dutch company that has a near-monopoly on manufacturing the machines used for chipmaking, may soon no longer be an isolated success story. Like its U.S. counterpart, the European Chips Act aims to foster the semiconductor industry — in part thanks to…

Google AI Studio is rolling out an ‘import from GitHub’ feature inside its Build mode. It takes a repo and transforms it into a runtime-compatible format. You can then keep iterating on it, deploy it, and more. The update was shared by the Google AI Studio account and by Logan…

National Highway Traffic Safety Administration (NHTSA) Administrator Jonathan Morrison issued a directive Wednesday to autonomous vehicle (AV) developers, stating that it is unacceptable for their vehicles to interfere with first responders or law enforcement. Morrison noted in the letter that the agency has “identified a clear pattern of driverless AVs interfering with law enforcement and…

SpaceXAI just released Grok 4.5. The company calls it its smartest model to date. It targets coding, agentic tasks, and knowledge work. SpaceXAI says Grok 4.5 was trained alongside Cursor, an AI coding editor. TL;DR Grok 4.5 targets coding, agentic tasks, and knowledge work, trained alongside Cursor. In SpaceXAI’s own chart, Fable (max) leads all…

Ant Group’s Robbyant has released LingBot-VLA 2.0, a Vision-Language-Action (VLA) foundation model for robots. The release includes a technical report, an Apache-2.0 codebase, and a 6B checkpoint. The research team targets a well-known gap: VLA models often work in labs but stumble in deployment. LingBot-VLA 2.0 advances the prior version…