RS-GQ-GC1-0109

BittWare
538-RS-GQ-GC1-0109
RS-GQ-GC1-0109

Fabricante:

Descripción:
Tarjetas aceleradoras GroqCard low latency AI/ML Inference PCIe accelerator card with single GroqChip

En existencias: 5

Existencias:
5 Se puede enviar inmediatamente
Mínimo: 1   Múltiples: 1
Precio unitario:
$ -
Precio ext.:
$ -
Est. Tarifa:

Precio (COP)

Cantidad Precio unitario
Precio ext.
$ 98.947.804 $ 98.947.804

Atributo del producto Valor de atributo Seleccionar atributo
Molex
Categoría de producto: Tarjetas aceleradoras
RoHS:  
Dual Width, Full Height, 3/4 Length PCI Express Gen4 x16 Adapter
230 MB
230 MB
Marca: BittWare
Empaquetado: Bulk
Tipo de producto: Accelerator Cards
Subcategoría: Embedded Solutions
TDP - Máx.: 275
Nombre comercial: Groq
Alias de las piezas n.º: RSGQGC10109
Productos encontrados:
Para mostrar productos similares, seleccione al menos una casilla de verificación
Seleccione al menos una de las casillas de verificación anteriores para mostrar productos similares en esta categoría.
Atributos seleccionados: 0

CAHTS:
8471801000
USHTS:
8471801000
JPHTS:
847180000
TARIC:
8471800000
MXHTS:
8471800402
BRHTS:
84718000
ECCN:
3A991.g

Acelerador GroqCard™

BittWare GroqCard™ Accelerator is a double-width PCIe form factor ML accelerator developed to integrate easily. The GroqWare™ suite implements a software-defined hardware approach, allowing easy deployment paths for PyTorch, TensorFlow, and ONNX-trained deep learning models. The BittWare GroqCard Accelerator features scalability with nine RealScale™ chip-to-chip connections that guarantee the deployment of multiple cards as efficiently as one. Furthermore, an internal software-defined network delivers predictable, repeatable performance with no run-to-run variations. The GroqCard has been qualified for use with the SMC AS-4124GS-TNR and Dell R750xa. The HPE DL385 Gen 10 Plus has been tested, but the full server interop exercise was not completed. In addition, liquid has also qualified the GroqCard in the chassis with up to 16 GroqCards. Using the GroqCard in other server models is at the user's risk.