Performance Rating
AMD Radeon Instinct MI250 outperforms GALAX GT 1030 by 5,256.84% in the overall GPU ARK performance rating
GALAX GT 1030
The AMD Radeon Instinct MI250 dramatically outperforms the GALAX GT 1030, leading by 5257% in the overall GPI performance index. For machine learning, the AMD Radeon Instinct MI250 has the edge in FP16 performance (+2057286%) and VRAM (128 vs 2 GB). For gaming, the AMD Radeon Instinct MI250 delivers 3915% more FP32 throughput. In performance per watt, the AMD Radeon Instinct MI250 is 141% more efficient at 500W TDP vs 30W for the GALAX GT 1030.
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks AdditionalMemory
Memory Size
+6,300%
128
ГБ
Memory Type
Memory Bandwidth
3.28 TB/s
Memory Bus Width
ML Performance
FP16 (Half Precision)
+2,057,286%
362.1
TFLOPS
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
+3,915%
45.26
TFLOPS
FP64 (Double Precision)
+128,480%
45.26
TFLOPS
CUDA Cores
+3,367%
13,312
RT Cores
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
PCIe Version
ML Software Support
CUDA Version
CUDA Toolkit (first supported)
CUDA Toolkit (deprecated)
CUDA Toolkit (support removed)
CUDA Toolkit status
Clocks & Performance
Base Clock
1,000
Boost Clock
+16%
1,700
Memory Clock
+7%
1,600
Power Consumption
Recommended PSU
-78%
200 W
Power Connector
TDP/TGP
-94%
30 W
Rendering
Texture Units (TMU)
+3,367%
832
ROP
L2 Cache
16 MB
Benchmarks
llama.cpp, llama-2-7b-Q4_0
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
