Nvidia Acquires Poolside in $6 Billion Deal to Expand AI Model Development

Nvidia has agreed to acquire AI startup Poolside in a $6 billion deal, signaling a major shift in its AI strategy. The agreement gives Nvidia access to Poolside’s technology and brings over 100 engineers from the startup into its workforce. Nvidia will also invest an additional $1 billion in Poolside, valuing the startup at $12 billion before the investment, according to The Wall Street Journal.

Key Highlights

• Nvidia acquires AI startup Poolside in a $6 billion deal.
• Over 100 Poolside engineers will join Nvidia’s Nemotron team.
• Nvidia invests an additional $1 billion at a $12 billion pre-money valuation.
• Rising memory chip costs may increase server prices by over 15 percent.
• Nvidia’s move intensifies competition with OpenAI and Anthropic.

Nvidia Expands AI Model Capabilities

The engineers from Poolside will join Nvidia’s Nemotron team, which focuses on developing open-weight AI models. These models allow users to download and modify them, offering more flexibility than closed systems from companies like OpenAI and Anthropic. Open-weight models can also reduce operational costs for businesses using AI.

The acquisition comes as Nvidia aims to strengthen the US position in the global AI race. Chinese companies have made significant progress in open AI models, especially after DeepSeek released a low-cost model in early 2025. Other Chinese firms, such as Moonshot and Z.AI, have also developed competitive models. Nvidia launched the Nemotron Coalition earlier this year, partnering with companies like Mistral, Thinking Machines Lab, and Perplexity to advance open-model development. Poolside’s addition brings more engineering talent and technology to this effort.

Impact on Industry and Market

Poolside was founded in 2023 by Eiso Kant and former GitHub CTO Jason Warner. The company recently introduced its Laguna S model and had previously struggled to secure enough funding for a large Nvidia-powered computing cluster. The new agreement provides Poolside with the computing resources needed for its long-term projects. While Poolside’s top leadership will remain independent, most of its engineers will transition to Nvidia.

This move places Nvidia in direct competition with some of its major customers, including OpenAI and Anthropic, as it develops open AI models. The expansion into AI models occurs as the cost of building AI infrastructure increases. According to Bloomberg, Nvidia has informed some of its largest customers that server prices containing its AI chips could rise by over 15% for systems shipped early next year. The affected systems use Nvidia’s Vera Rubin and Grace Blackwell platforms. The main factor behind the price increase is the rising cost of memory chips, with Samsung, SK Hynix, and Micron dominating the DRAM market. High demand from AI data centers has tightened supply and driven up prices.

Nvidia now faces the challenge of making its AI models more competitive while managing higher hardware costs. The company is set to report its quarterly results on August 26. Investors are watching closely to see how these infrastructure costs will affect Nvidia and the broader AI sector.