Ampere GA100 · 24 GB HBM2 · 933 GB/s · 165 W
Ampere (GA100)
3,584
165 TFLOPS
82 TFLOPS
24 GB HBM2
933 GB/s
165 W
PCIe Gen4 x16
IN STOCK
Quote
The NVIDIA A30 is a datacenter Tensor Core GPU built on the GA100 die, with 24 GB of HBM2 and multi-instance GPU (MIG) partitioning for consolidated inference and mainstream training. QS Compute supplies A30 boards for AI inference servers and virtualized datacenter deployments.
Need NVIDIA A30 24GB?
NVIDIA A30 24GB available. MIG-capable Ampere datacenter GPU for AI inference.
Request Quote