NVIDIA
NVIDIA RTX PRO 6000 Blackwell 96GB
Blackwell professional card with 96 GB GDDR7 ECC, offered in workstation, Max-Q workstation and server editions.
| Dostępność | Przedsprzedaż |
|---|---|
| SKU | AXL-GPU-RTXPRO6000 |
| Minimalne zamówienie | 1 unit |
| Stan | Nowy |
| Ceny | Na zapytanie |
Opis
The RTX PRO 6000 Blackwell moves the professional line to 96 GB of GDDR7 with ECC on a 512-bit interface, enough memory to keep sizeable models, scenes or datasets local. It comes in three editions: a Workstation Edition, a lower-power Max-Q Workstation Edition for multi-card towers and quiet studios, and a passively cooled Server Edition for rack deployment. Pick the edition against your chassis power and airflow budget, because they are not interchangeable.
AXISLINK confirms the exact edition, cooling type and part number on the quotation along with the serial number of each board. We ship cards individually or inside validated workstation and server builds where power delivery and thermals have been tested at the intended card count. Export documentation and freight from Hong Kong are arranged with the order, and end-user information may be requested for higher-tier accelerators.
Specyfikacja
| Architecture | NVIDIA Blackwell |
|---|---|
| Memory | 96 GB GDDR7 with ECC |
| Memory interface | 512-bit |
| Editions | Workstation Edition, Max-Q Workstation Edition, Server Edition |
|---|---|
| Form factor | Dual-slot; workstation editions actively cooled, Server Edition passively cooled |
| Display outputs | 4x DisplayPort 2.1 on workstation editions, none on Server Edition |
| Host interface | PCIe 5.0 x16 |
| Board power | Up to 600 W on Workstation and Server editions, 300 W on Max-Q Workstation Edition |
|---|---|
| Cooling | Active blower on workstation editions, chassis airflow on Server Edition |
Zapytaj o wycenę
Wycena dla NVIDIA RTX PRO 6000 Blackwell 96GB. Odpowiadamy formalną fakturą proforma, zwykle w ciągu 24 godzin.
Więcej w kategorii Karty GPU i akceleratory
Zobacz kategorię
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.