NVIDIA AI Updates: June 30, 2026

1. Anthropic’s Claude Models Now Run on NVIDIA GB300 in Microsoft Azure

NVIDIA. NVIDIA announced that Anthropic’s Claude models are now running on NVIDIA GB300 NVL72 systems based on the Blackwell Ultra architecture in Microsoft Azure. The deployment marks a deepening of the relationship between Anthropic, NVIDIA, and Microsoft, bringing Claude inference to NVIDIA’s latest rack-scale platform. NVIDIA positions GB300 as delivering higher throughput and efficiency for large-scale model serving. Source

2. Firefly Aerospace Operates NVIDIA Jetson in Lunar Orbit for the First Time

NVIDIA. NVIDIA said Firefly Aerospace operated an NVIDIA Jetson module in lunar orbit for the first time, marking a milestone for edge AI computing in space. The Jetson platform enables onboard processing for spacecraft, reducing reliance on ground-based computation. The deployment demonstrates ruggedized accelerated computing operating in the harsh conditions of cislunar space. Source

3. Palantir Brings Secure AI to US Agencies With NVIDIA Nemotron

NVIDIA. NVIDIA detailed how Palantir is bringing open Nemotron models into closed, secure environments to deliver AI for US government agencies. The collaboration pairs NVIDIA’s open Nemotron model family with Palantir’s platforms to run AI workloads inside air-gapped and restricted settings. The approach aims to give agencies access to open models while maintaining security and data control. Source

4. NVIDIA Outlines How to Govern Autonomous Agents in Enterprise AI Factories

NVIDIA. NVIDIA published guidance on governing autonomous AI agents in enterprise AI factories, noting agents increasingly inspect code, run tests, read documents, query internal systems, and operate for hours at a time. The post describes approaches for securing and overseeing long-running agentic workloads in production. It addresses the operational and security challenges of moving agents beyond chat into enterprise systems. Source