Performance Rating
GUNNIR Arc B580 Photon OC outperforms NVIDIA A2 by 123.74% in the overall GPU ARK performance rating
NVIDIA A2
The GUNNIR Arc B580 Photon OC dramatically outperforms the NVIDIA A2, leading by 124% in the overall GPI performance index. For machine learning, the GUNNIR Arc B580 Photon OC has the edge in FP16 performance (+544%). For gaming, the GUNNIR Arc B580 Photon OC delivers 222% more FP32 throughput, and the GUNNIR Arc B580 Photon OC has 100% more RT cores. The NVIDIA A2 consumes less power (60W vs 190W TDP).
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering AdditionalMemory
Memory Size
+33%
16
ГБ
Memory Type
Memory Bandwidth
Memory Bus Width
ML Performance
FP16 (Half Precision)
+544%
29.18
TFLOPS
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
+222%
14.59
TFLOPS
FP64 (Double Precision)
+1,188%
0.912
TFLOPS
CUDA Cores
+100%
2,560
RT Cores
+100%
20
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
10
PCIe Version
ML Software Support
CUDA Version
CUDA Toolkit (first supported)
CUDA Toolkit status
Clocks & Performance
Base Clock
+85%
2,670
Boost Clock
+61%
2,850
Memory Clock
+52%
2,375
Power Consumption
Recommended PSU
-44%
250 W
Power Connector
TDP/TGP
-68%
60 W
Rendering
Texture Units (TMU)
+300%
160
ROP
+100%
20
L2 Cache
2 MB
Additional
Slots
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
