NVIDIA & Thinking Machines Labs partner
NVIDIA partnered with Thinking Machines Labs (an OpenAI rival) to deliver a gigawatt of AI chips and customizable AI at scale.
The partnership will focus on deploying NVIDIA's Vera Rubin AI systems to bolster Thinking Machines Lab's AI model training. This infrastructure deployment is slated to begin early next year. The collaboration aims to broaden access to advanced AI models for enterprises, research institutions, and the scientific community. NVIDIA's investment in Thinking Machines Lab will support the company's long-term growth. The partnership will also focus on designing training and serving systems optimized for NVIDIA hardware. Mira Murati, CEO of Thinking Machines, emphasized that NVIDIA's technology is the foundation of the AI field. Thinking Machines Lab, founded by Murati in February 2025, has already raised over $2 billion. Investors include Andreessen Horowitz, Accel, and AMD's venture arm, alongside NVIDIA. The company's headcount has grown from roughly 30 to around 120 employees in the past year. Thinking Machines Lab's primary focus is on customizable AI, with an emphasis on human-AI collaboration. Their flagship product, Tinker, is a training API designed for fine-tuning open-source models. The company aims to make AI more understandable and adaptable to various fields.
