Thinking Machines Gets NVDA Investment

$NVDA is investing in Thinking Machines Lab, supplying Vera Rubin AI chips with 1GW compute online early 2027 [https://x.com/i/status/2031357111633064424]. Traders are bearish on $SPY/$SPX amid VIX spikes, with predictions of drops [https://x.com/i/status/2031360532322656662]. What is the Vera Rubin AI chip designed to do?

Thinking Machines Lab, founded in February 2025 by ex-OpenAI CTO Mira Murati, is receiving a substantial investment from Nvidia. The deal includes access to at least 1 gigawatt of Nvidia's next-generation Vera Rubin systems, slated for deployment in early 2027. This level of computing power can cost around $50 billion and is enough to power roughly 750,000 U.S. homes. The partnership aims to develop training and serving systems optimized for Nvidia architectures, expanding access to frontier AI and open models. Thinking Machines Lab secured $2 billion in seed funding led by Andreessen Horowitz in July 2025, valuing the company at $12 billion. Investors included Nvidia, AMD, Cisco, and Jane Street. The Vera Rubin platform, named after astronomer Vera Florence Cooper Rubin, features six new chips designed for an AI supercomputer. It unifies 72 Rubin GPUs, 36 Vera CPUs, ConnectX-9 SuperNICs, and BlueField-4 DPUs. The platform is built for agentic AI and reasoning, reducing inference token costs by up to 10x compared to the Nvidia Blackwell platform. The Vera Rubin NVL72 rack-scale system delivers 3.6 exaflops of FP4 inference compute, significantly faster than previous generations. Each Rubin GPU boasts 288GB of HBM4 memory, offering 2.8x the memory bandwidth of Blackwell. Nvidia has already begun shipping Vera Rubin samples to select customers. Meanwhile, bearish sentiment surrounds the broader market, with the VIX spiking and predictions of drops in the SPY/SPX. The Cboe Volatility Index ($VIX) reached its highest point since October 2025. Despite Nvidia's earnings beat in late 2025, SPX struggled to shake off bearish flows. Thinking Machines Lab's founding team included OpenAI co-founder John Schulman, former VP of Research Barret Zoph, and Lilian Weng. The company launched Tinker, an API for fine-tuning language models, in October 2025. Tinker allows users to submit jobs for fine-tuning open-weight models on Thinking Machines Lab's internal clusters. Mira Murati's departure from OpenAI in 2024 followed the brief "Sam Altman ouster". Thinking Machines Lab is structured as a public benefit corporation, with Murati holding a deciding vote on board matters. Founding shareholders possess votes weighted 100 times greater than those of regular shareholders. Nvidia's investment underscores its growing role in financing AI startups that rely on its chips. Thinking Machines Lab aims to democratize access to AI knowledge and tools, enabling customization for specific needs. The company's mission is to build collaborative general intelligence, emphasizing open science, transparency, and human-AI partnerships.

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.