NVIDIA designs its AI GPUs, but it relies on specialist suppliers to fabricate wafers, package chips, provide memory, and assemble and test finished products. Its fiscal 2026 Form 10-K names TSMC and Samsung as wafer foundries, says NVIDIA uses CoWoS for semiconductor packaging, and identifies SK hynix, Micron, and Samsung as memory sources. That makes advanced packaging and high-bandwidth memory integral parts of the production chain—not optional add-ons to a GPU die.
Why an AI GPU depends on more than its processor die
A high-performance AI GPU is a system of tightly integrated components. The logic die performs computation, while high-bandwidth memory (HBM) supplies data at the rates needed to keep that compute capacity fed. Advanced packaging brings dies and memory together at high density, making packaging part of how the overall system is built.
This creates a multi-stage supply chain: wafer fabrication produces semiconductor dies; advanced packaging integrates components; memory suppliers provide memory; and contract manufacturers perform assembly, testing, and final-product packaging. A shortage or delay at any stage can affect the pace at which complete systems are produced, although NVIDIA’s public disclosures do not identify one stage as the limiting factor for a particular GPU.
Which suppliers NVIDIA names, and what they do
| Production stage | Companies NVIDIA identifies | What the disclosure establishes |
|---|---|---|
| Wafer fabrication | TSMC and Samsung | NVIDIA’s fiscal 2026 Form 10-K identifies them as foundries producing its semiconductor wafers. |
| Advanced semiconductor packaging | CoWoS technology | NVIDIA says it uses CoWoS for semiconductor packaging. The filing does not quantify capacity or assign a share of packaging to a particular GPU model. |
| Memory | SK hynix, Micron, and Samsung | NVIDIA names these companies as memory sources. It does not disclose a model-by-model supplier allocation or say that all three supply every GPU generation. |
| Assembly, testing, and final-product packaging | Hon Hai, Wistron, and Fabrinet | NVIDIA identifies these as examples of independent subcontractors and contract manufacturers involved in this work. |
The table describes different roles, not interchangeable suppliers or a ranking. A company named at one stage should not be assumed to supply another stage for every product.
#1 Best Overall
- System Compatibility Note: This 2-slot card measures 271 x 112 x 39 mm and requires a single 12V-2x6-pin power connector. Please verify chassis and PSU compatibility before purchase.
- Dedicated Support: Please contact us directly through Amazon for any product questions or assistance you may require.
- Professional Intel Arc Pro B70 GPU: Built on the Intel Xe2-HPG architecture, it features 32 Xe cores and 256 XMX engines, designed to accelerate AI, rendering, and complex visualization workloads.
- Massive 32GB GDDR6 VRAM: Equipped with 32GB of high-speed GDDR6 memory on a 256-bit bus, running at 19 Gbps, which allows for handling large AI models and complex datasets locally.
- High-Performance Engine Clock: Delivers an engine clock of 2540 MHz, providing the compute power needed for demanding professional applications and AI inference.
What CoWoS packaging contributes
CoWoS is a semiconductor packaging technology NVIDIA says it uses. In the broader system, advanced packaging enables chips and memory to be integrated at high density, helping form a tightly connected package for demanding workloads. The company’s cited filing confirms the technology’s use but does not give a per-GPU package design, CoWoS capacity figure, cost allocation, or supplier share.
That distinction matters when discussing supply constraints. Packaging is clearly a dependency in NVIDIA’s production chain, but the available disclosures do not establish that CoWoS alone is the bottleneck for a specific product or at a specific time.
Rank #2
- NVIDIA Volta GV100 Architecture — 4,608 CUDA Cores, 640 1st-Gen Tensor Cores delivering 14 TFLOPS FP32 and 112 TFLOPS deep learning performance for AI training, inference, HPC, and scientific computing workloads
- 32GB HBM2 ECC Memory — 900 GB/s Bandwidth — High-bandwidth memory on a 4096-bit bus with ECC error correction provides the memory capacity and throughput required for the largest AI models, simulations, and datasets
- PCIe 3.0 x16 Interface — 250W TDP — Standard PCIe Gen3 connectivity with passive cooling designed for enterprise rack server deployment in HPE ProLiant, Dell PowerEdge, and Supermicro platforms with adequate chassis airflow
- NVLink — Scale to 96GB Unified Memory — Connect two V100 GPUs via NVLink at 300 GB/s bi-directional bandwidth to scale GPU memory from 32GB to 96GB for larger AI training and HPC workloads
- Multi-Precision Computing — Supports FP64 (7 TFLOPS), FP32 (14 TFLOPS), FP16 (112 TFLOPS) and INT8 precision modes for flexible deployment across training, inference, and scientific simulation workloads
Who makes the memory in NVIDIA AI GPUs?
NVIDIA’s fiscal 2026 Form 10-K names SK hynix, Micron, and Samsung as memory sources. It does not break down which company supplies memory for each GPU model, how much each provides, or whether the three suppliers participate equally. A supplier list is evidence of sourcing relationships, not a per-product bill of materials.
SK hynix: a public next-generation partnership
On June 7, 2026, NVIDIA announced a multiyear partnership with SK hynix to advance next-generation memory in line with NVIDIA’s infrastructure roadmap. NVIDIA said the work includes memory for Vera Rubin AI supercomputers and other platforms. The announcement shows collaboration and roadmap alignment; it does not disclose shipment shares by supplier or model.
Rank #3
- Professional AI & Creator Workstation: AMD Radeon AI PRO R9700 GPU with 32GB GDDR6 is engineered for AI development, professional content creation, and compute-intensive workloads.
- Massive 32GB Memory Capacity: 32GB of GDDR6 memory on a 256-bit bus provides ample bandwidth for large AI models, 8K video editing, and complex 3D rendering.
- Advanced RDNA 4 with AI Accelerators: 64 Compute Units with 3rd Gen Ray Tracing and dedicated 2nd Gen AI Accelerators for groundbreaking AI performance and visual computing.
- Professional Blower Cooling: Efficient single blower design exhausts heat directly out of the chassis, ideal for multi-GPU workstation and server configurations.
- Enterprise-Grade Thermal Solution: Vapor chamber heatsink with industrial Honeywell PTM7950 thermal interface material ensures reliable cooling under sustained professional loads.
Samsung: memory, foundry, and design collaboration
NVIDIA’s Samsung announcement describes work spanning HBM3E and HBM4, memory, foundry services, chip design, computational lithography, and factory operations. NVIDIA reported 20x performance gains for specified computational-lithography and technology-computer-aided-design simulations in that collaboration. This is a company-reported result for those workflows, not a GPU performance claim or an independently verified manufacturing outcome.
Does NVIDIA manufacture its own AI chips?
NVIDIA designs its GPUs, but its fiscal 2026 Form 10-K describes external foundries producing its semiconductor wafers. In that sense, NVIDIA is fabless for the wafer production described in the filing: it depends on partners for fabrication rather than making those wafers in its own foundries. Other partners contribute memory, packaging, assembly, and testing, so “who makes the GPU?” has different answers depending on which stage is meant.
Rank #4
- System Compatibility Note: 2-slot card, 271x112x39mm, single 8-pin power, 200W TDP. Verify chassis clearance and PSU capacity before purchase.
- Dedicated Support: Please contact us directly through Amazon for any product questions or assistance you may require.
- 24GB GDDR6 on 192-Bit Bus: Massive 24GB memory with 456 GB/s bandwidth – ideal for LLMs, AI inference, 3D rendering, and generative design.
- Intel Xe2-HPG Architecture: Built on Intel's next-gen architecture with 20 Xe cores and 160 XMX engines for AI acceleration (197 INT8 TOPS).
- PCIe 5.0 Support: PCI Express 5.0 x16 interface for maximum bandwidth with the latest workstation platforms.
NVIDIA says its supply chain was mainly concentrated in Asia and that it was expanding into the United States and Latin America. The filing cautions that scaling in new locations depends on local ecosystems reaching required volumes on time; expansion should not be read as proof that production has already shifted at a particular scale.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What NVIDIA’s latest disclosed supply commitments mean
In its Form 10-Q for the quarter ended July 26, 2026, NVIDIA reported supply and capacity commitments rising from $119 billion in the prior quarter to $279 billion. The company described these broad commitments as primarily for memory and manufacturing facilities, in connection with data-center infrastructure systems. The figures are not an HBM-only, CoWoS-only, or single-GPU amount, and the filing does not apportion them among suppliers or technologies.
Best Value
- PLEASE NOTE: Exporting an NVIDIA RTX Pro 6000 GPU outside the US requires strict adherence to the U.S. Export Administration Regulations (EAR) and issuance of an export license from the Bureau of Industry and Security (BIS). Compliance and Know Your Customer (KYC) screening may be required as a condition of order acceptance. [NVIDIA Blackwell Streaming Multiprocessor] The new SM features increased processing throughput, and new neural shaders that integrate neural networks inside of programmable shaders | DLSS 4: Multi Frame Generation ensures ultra-smooth frame pacing for lifelike simulations.
- [Double-Flow-Through Design] The RTX PRO 6000 Blackwell features a double-flow-through cooling design, optimizing efficiency and airflow to sustain peak performance under 600W power loads. | [5th Gen Tensor Cores] Deliver up to 3X the performance of the previous generation and support for FP4 precision for faster AI model processing times with reduced memory usage, enabling local fine-tuning of LLMs and generative AI | [4th Gen Ray Tracing Cores] Double the ray-triangle intersection rate of the previous generation to create photoreal, physically accurate scenes and immersive 3D designs with RTX Mega Geometry, which enables up to 100X more ray-traced triangles.
- [PCIe Gen 5] Support for PCIe Gen 5 provides double the bandwidth of PCIe Gen 4, improving data-transfer speeds from CPU memory and unlocking faster performance for data-intensive tasks like AI, data science, and 3D modeling. | [GDDR7 Memory] With 96 GB of GPU memory and 1.8 TB ps bandwidth, it can tackle massive 3D and AI projects, fine-tune AI models locally, explore large-scale VR environments, and drive larger multi-app workflows.
- [DisplayPort 2.1] Achieve unparalleled visual clarity and performance, driving high resolution displays at up to 8K at 240 Hz and 16K at 60 Hz. Increased bandwidth enables seamless multi-monitor setups while HDR and higher color depth support ensures superior color accuracy for precision work, such as video editing, 3D design, and live broadcasting.
- [Universal MIG] Divide a single RTX PRO 6000 Blackwell into multiple isolated instances, each with dedicated resources, allowing for concurrent execution of multiple workloads, optimized GPU utilization, and secure isolation of different applications or users. [WARRANTY] 3 YR Manufacturer's Warranty. Bulk OEM Packaging. Retail Packaging is NOT included.
The same filing said Blackwell accounted for the majority of system shipments, Vera Rubin had begun production shipments, and NVIDIA was experiencing certain supply constraints. These are statements about conditions reported for that filing period, not permanent descriptions of the product mix or a diagnosis of which supplier or production stage was constrained.
Quick Recap
What the public disclosures do—and do not—show
- Established: NVIDIA names wafer foundries, memory sources, and contract manufacturers, and says it uses CoWoS packaging.
- Not established: the portion of a given GPU’s supply or cost attributable to packaging, memory, or an individual supplier.
- Not established: supplier-by-supplier market share, allocation, capacity, yield, or which supplier is the bottleneck for a particular GPU.
- Time-sensitive: shipment mix, supply commitments, production status, and supply constraints can change; the cited operational statements are dated to NVIDIA’s Form 10-Q for the quarter ended July 26, 2026.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
