ELSA JAPAN Inc. (Est. 1997) is a leading Japanese distributor of NVIDIA professional and consumer GPUs, Supermicro server solutions, and cutting-edge AI computing platforms. Based on computer graphics and multimedia technology, ELSA JAPAN partners with world-class vendors to deliver optimal GPU and AI products to the Japanese market.
Our Products
NVIDIA GPU
NVIDIA RTX PRO 6000 Blackwell Workstation Edition
Professional GPU for AI & Visualization
The latest professional-grade GPU built on NVIDIA Blackwell architecture. Designed for AI development, scientific simulation, 3D visualization, and generative AI workloads. Massive VRAM, RT Cores, and Tensor Cores deliver unmatched performance for engineers, researchers, and content creators.
Blackwell Architecture / RT Cores / Tensor Cores / Professional Drivers / ECC Memory
NVIDIA GeForce RTX 50 Series
Next-Gen Consumer GPUs (INNO3D)
The full lineup of NVIDIA GeForce RTX 50 series through INNO3D, featuring RTX 5090, 5080, 5070 Ti, 5070, 5060 Ti, 5060, and 5050. iCHILL FROSTBITE, X3 OC, and TWIN X2 cooling designs for gaming, content creation, and AI-powered applications.
RTX 5090 / 5080 / 5070 Ti / 5070 / 5060 Ti / 5060 / 5050 / INNO3D / Blackwell
NVIDIA Data Center GPUs
H200 NVL & Enterprise AI Accelerators
Enterprise-grade data center GPUs including NVIDIA H200 NVL with 141GB HBM3e memory for large-scale AI training, inference, and HPC workloads. Designed for AI factories, cloud data centers, and research institutions requiring maximum throughput and scalability.
H200 NVL / HBM3e / NVLink / AI Training / Inference / HPC
Supermicro Servers
Supermicro GPU Workstations
Professional AI & Visualization Workstations
High-performance workstations powered by Supermicro hardware, optimized for multi-GPU configurations. Ideal for AI development, deep learning model training, 3D rendering, and CAD/CAE simulations. ELSA JAPAN has been an authorized Supermicro partner since 2012.
Multi-GPU / Intel Xeon / AMD EPYC / ECC Memory / Professional Grade / ISV Certified
Supermicro Rack-Mount GPU Servers
Data Center AI Infrastructure
Enterprise rack-mount servers designed for AI/ML training and inference at scale. Supporting multiple NVIDIA GPUs with high-speed NVLink interconnects, massive storage, and redundant power supplies. Ready for deployment in AI factories, research labs, and cloud service providers.
1U / 2U / 4U / Multi-GPU / NVLink / Redundant PSU / IPMI / Data Center Ready
DGX Spark
NVIDIA DGX Spark
Desktop AI Supercomputer
A personal AI supercomputer powered by the NVIDIA GB10 Grace Blackwell Superchip. Compact desktop form factor with up to 128GB unified memory for prototyping, fine-tuning, and running large AI models locally. NVIDIA DGX Cloud integration for seamless scale-out to data center infrastructure.
GB10 Grace Blackwell / 128GB Unified Memory / Desktop Form Factor / DGX Cloud Ready
DGX Spark Rack Mount Kit (1.5U / 2-Unit)
ELSA Original 19-inch Rack Accessory
ELSA JAPAN's proprietary 19-inch 1.5U rack mount kit designed to house two NVIDIA DGX Spark units in a standard server rack. Enables compact, organized deployment of multiple DGX Spark systems for research labs and development teams requiring scalable local AI compute.
1.5U / 19-inch Rack / 2x DGX Spark / ELSA Original / Data Center Deployment