Amazon (AMZN) plans to use Nvidia’s (NVDA) latest NVLink Fusion technology in its next generation artificial intelligence (AI) accelerators, Trainium4. Amazon Web Services (AWS) announced the plan at its annual re:Invent conference in Las Vegas. NVLink Fusion enables faster, higher-bandwidth connections across multiple chips, which is crucial for training large AI models, potentially lowering training time and costs for customers.
TipRanks Cyber Monday Sale
- Claim 60% off TipRanks Premium for data-backed insights and research tools you need to invest with confidence.
- Subscribe to TipRanks' Smart Investor Picks and see our data in action through our high-performing model portfolio - now also 60% off
AWS has been expanding its efforts to attract more AI customers to use its services. At the same time, Nvidia is pursuing broader NVLink adoption beyond its own platforms, with Intel (INTC) and Qualcomm (QCOM) already on board.
More Details of the AMZN-NVDA Partnership
NVLink Fusion will enable AWS to build larger AI servers that recognize and communicate more quickly, a key requirement for training large AI models that link thousands of machines together. Through the Nvidia partnership, AWS customers will also gain access to AI Factories, exclusive AI infrastructure embedded in their own data centers to boost speed and readiness.
Nvidia CEO Jensen Huang stated that the two companies are building hardware and cloud tools to make AI mainstream in every company and country, accelerating progress toward smarter systems.
AWS Rolls Out New Chip and Tools
At the conference, AWS launched Trainium3, an in-house AI chip built on 3-nanometer tech, along with Trainium3 UltraServers. The chip aims for faster AI training and deployment, lower costs, and better energy efficiency. It competes directly with Nvidia and Alphabet (GOOGL) in the AI chip market.
Furthermore, AWS introduced Nova Forge, a tool for customizing AI models with proprietary data early in training, priced at $100,000 per year. Unlike post-training fine-tuning, Nova Forge integrates data during model development. Amazon claims Nova Forge is far cheaper than building models from scratch, which run into hundreds of millions or billions of dollars.
AMZN or NVDA: Which Is the Better AI Stock?
According to TipRanks’ Stock Comparison tool, analysts currently have a “Strong Buy” consensus rating on both AMZN and NVDA shares. However, AMZN stock boasts a TipRanks Smart Score of Perfect 10, which implies Amazon is highly likely to outperform market expectations. Meanwhile, NVDA offers a higher upside potential over the next twelve months.
