NVIDIA
NVIDIA GeForce RTX 4090
Ada Lovelace card with 24 GB GDDR6X, sourced from the secondary market for render farms and workstation builds.
| 在庫状況 | 在庫あり |
|---|---|
| SKU | AXL-GPU-4090 |
| 最低注文数量 | 4 units |
| 状態 | 整備済み |
| 価格 | 見積りにて |
概要
The GeForce RTX 4090 held the top of the Ada consumer line and is still wanted for its 24 GB of GDDR6X, more memory than the current generation offers at the same tier. It remains a common choice for offline rendering, video and 3D production, simulation front-ends and local model experimentation.
New retail supply of this model has largely dried up, so AXISLINK sources verified refurbished cards instead of promising new stock we cannot secure. Every unit is cleaned, functionally tested and run under load, the partner model and serial number appear on the quotation, and the condition grade is stated in writing. Cards ship loose in protective packaging or built into a tested workstation, with export documentation and freight arranged from Hong Kong.
仕様
| Architecture | NVIDIA Ada Lovelace |
|---|---|
| Memory | 24 GB GDDR6X, 384-bit |
| Memory bandwidth | 1008 GB/s |
| CUDA cores | 16384 |
| Form factor | PCIe 4.0 x16 add-in card, triple-slot on most partner designs |
|---|---|
| Display outputs | DisplayPort 1.4a and HDMI 2.1, layout varies by partner |
| Total graphics power | 450 W |
|---|---|
| Connector | 16-pin 12VHPWR |
見積依頼
NVIDIA GeForce RTX 4090のお見積りです。通常24時間以内に正式なプロフォーマインボイス（見積送り状）でご返信します。
GPU・アクセラレータのその他の製品
カテゴリーを見る
NVIDIA
NVIDIA H200 141GB SXM
Hopper SXM module with 141 GB HBM3e for large-model training and memory-bound inference on HGX baseboards.
NVIDIA
NVIDIA L40S 48GB
Ada Lovelace data-center card with 48 GB GDDR6 ECC for inference, fine-tuning, rendering and virtual workstations.
NVIDIA
NVIDIA H100 80GB
Hopper-architecture data-center GPU for large-scale AI training and inference. PCIe and SXM variants.