Samsung to Use TSMC's 3nm Process to Manufacture Next-Generation HBM Base Dies for NVIDIA
Samsung will use TSMC's 3nm process to manufacture next-generation HBM base dies for NVIDIA. NVIDIA has also required SK Hynix and Micron to adopt the same process, breaking Samsung's previous fully in-house manufacturing model.
Key Takeaways
- Key Highlight：Samsung will use TSMC's 3nm process to manufacture next-generation HBM base dies for NVIDIA. NVIDIA has also required SK Hynix and Micron to adopt the same process, breaking Samsung's previous fully in-house manufacturing model.
- Innovation & Tech：Highlights advancements in NVIDIA, Samsung, Use, demonstrating rapid progress in model capabilities.
- Industry Impact：Reported via IT之家 (CN), offering actionable signals for developers and technology leaders.
Samsung Electronics will use TSMC's 3nm process to manufacture the base dies for NVIDIA's next-generation High Bandwidth Memory (HBM). Previously, Samsung produced both the core dies and base dies for its HBM products in-house. This shift to TSMC foundry services marks a significant adjustment to its manufacturing strategy.
NVIDIA has issued a unified requirement to its three HBM suppliers—Samsung, SK Hynix, and Micron—to use TSMC's 3nm process for manufacturing base dies. HBM is a critical component of AI training and inference chips, and NVIDIA GPUs are highly dependent on HBM bandwidth and energy efficiency. This requirement directly serves the performance upgrade of AI computing chips.
Samsung's decision to abandon in-house manufacturing of base dies reflects the challenges facing the IDM model at advanced process nodes. TSMC's 3nm process offers advantages in power consumption and integration density, making it the optimal choice for memory paired with AI accelerators.
This supply chain adjustment means that NVIDIA is further strengthening its unified control over upstream processes in its next-generation AI chips. The three HBM suppliers will share the same manufacturing platform for base dies, which may affect each company's room for differentiation in HBM competition.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.
Industry Insights & Analysis
As artificial intelligence rapidly evolves, breakthroughs surrounding NVIDIA, Samsung, Use, TSMC are shifting toward scalable, robust real-world implementations.
Driven by both open-source ecosystems and proprietary model architectures, the integration between compute optimization, data engineering, and agentic workflows is accelerating. This development provides a strategic benchmark for upcoming AI tooling and developer workflows.