A30

Graphics processor

Video memory: 24 GB^*
Memory bus: 3,072 bit^*
Rated memory bandwidth: 933 GB/s^*

Benchmark results

Geekbench GPU (reported compilation) · version unreported · CUDA139,688^*Platform unreported Compare results
Geekbench GPU (reported compilation) · version unreported · CUDA139,688^*Platform unreported Compare results
Geekbench GPU (reported compilation) · version unreported · OpenCL118,447^*Platform unreported Compare results
Geekbench GPU (reported compilation) · version unreported · OpenCL118,447^*Platform unreported Compare results

Processing & platform

Architecture: Ampere^*

Memory & storage

Video memory: 24 GB^*
Video memory type: HBM2^*
Memory bus: 3,072 bit^*
Rated memory bandwidth: 933 GB/s^*

System configuration

Detailed specifications

Specifications 1 · 20 fields

GPU architecture — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: Ampere^*
Theoretical peak: BF16 tensor dense per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 165^* TFLOPS
Theoretical peak: FP16 tensor dense per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 165^* TFLOPS
Theoretical peak: FP32 non-tensor dense per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 10.3^* TFLOPS
Native hardware support: FP4 tensor — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: No^*
Theoretical peak: FP64 non-tensor dense per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 5.2^* TFLOPS
Theoretical peak: FP64 tensor dense per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 10.3^* TFLOPS
Native hardware support: FP6 tensor — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: No^*
Native hardware support: FP8 tensor — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: No^*
Theoretical peak: INT8 tensor dense per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 330^* TOPS
Theoretical peak: TF32 tensor dense per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 82^* TFLOPS
Physical configuration — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: PCIe, dual-slot FHFL^*
GPU interconnect bandwidth (NVLink 3) — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 200^* GB/s
PCIe interface declaration — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: PCIe Gen4 x16^*
Maximum board power — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 165^* W
Peak memory bandwidth per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 933^* GB/s
Memory interface width per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 3,072^* bit
Memory capacity per GPU — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: 24^* GB
Memory technology — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: HBM2^*
GPU model — NVIDIA A30 Tensor Core GPU — 24 GB HBM2: NVIDIA A30 Tensor Core GPU^*

Specifications 2 · 4 fields

Manufacturer

Nvidia^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Device

A30^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

CUDA

139,688^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

OpenCL

118,447^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Specifications 3 · 4 fields

Manufacturer

Nvidia^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Device

A30^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

CUDA

139,688^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

OpenCL

118,447^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

* Provisional data