Performance Rating
NVIDIA A800 SXM4 80 GB outperforms NVIDIA A2 by 553.79% in the overall GPU ARK performance rating
NVIDIA A800 SXM4 80 GB
Expert Comparison
NVIDIA A2 лучше для базовых задач графики и轻量级应用，如视频播放、简单的图形设计和轻度游戏。它具有16GB GDDR6内存和60W TDP，适合能耗控制严格的环境。相比之下，NVIDIA A800 SXM4 80 GB 更适合高性能计算和深度学习任务，拥有80GB HBM2e高速内存和400W TDP，性能远超A2，是数据中心和专业工作站的理想选择。
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering AdditionalMemory
Memory Size
+400%
80
ГБ
Memory Type
Memory Bandwidth
2.04 TB/s
Memory Bus Width
ML Performance
FP16 (Half Precision)
+1,621%
77.97
TFLOPS
BF16 (Brain Float)
311.84
TFLOPS
TF32 (TensorFloat)
155.92
Compute Power
FP32 (Single Precision)
+330%
19.49
TFLOPS
FP64 (Double Precision)
+13,666%
9.746
TFLOPS
CUDA Cores
+440%
6,912
RT Cores
10
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
+980%
108
PCIe Version
ML Software Support
CUDA Version
8.6
CUDA Toolkit (first supported)
CUDA Toolkit status
Clocks & Performance
Base Clock
+25%
1,440
Boost Clock
+26%
1,770
Memory Clock
+2%
1,593
Power Consumption
Recommended PSU
-69%
250 W
Power Connector
TDP/TGP
-85%
60 W
Rendering
Texture Units (TMU)
+980%
432
ROP
10
L2 Cache
+1,900%
40 MB
Additional
Slots
SXM Module
Release Date
Display Outputs
Renting is cheaper than buying
H100 80Gb SXM (80GB)
NVIDIA
Флагманская GPU на архитектуре Hopper с 80 ГБ памяти HBM3 в форм-факторе SXM. Создана для обучения и инференса больших языковых моделей, HPC задач. Поддерживает FP64, FP32, TF32, FP16, BF16, FP8 и INT8 вычисления с Tensor Cores 4-го поколения.
from 299.11 RUB/hour
