Performance Rating
AMD Radeon Instinct MI325X outperforms Intel Arc A570M by 2,406.27% in the overall GPU ARK performance rating
Intel Arc A570M
The AMD Radeon Instinct MI325X dramatically outperforms the Intel Arc A570M, leading by 2406% in the overall GPI performance index. For machine learning, the AMD Radeon Instinct MI325X has the edge in FP16 performance (+6038%) and VRAM (288 vs 8 GB). For gaming, the AMD Radeon Instinct MI325X delivers 1435% more FP32 throughput, and the Intel Arc A570M supports hardware ray tracing. In performance per watt, the AMD Radeon Instinct MI325X is 15% more efficient at 1000W TDP vs 75W for the Intel Arc A570M.
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks AdditionalMemory
Memory Size
+3,500%
288
ГБ
Memory Type
Memory Bandwidth
10.3 TB/s
Memory Bus Width
ML Performance
FP16 (Half Precision)
+6,038%
653.7
TFLOPS
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
+1,435%
81.72
TFLOPS
FP64 (Double Precision)
81.72
TFLOPS
CUDA Cores
+850%
19,456
RT Cores
16
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
PCIe Version
ML Software Support
CUDA Version
Clocks & Performance
Base Clock
+11%
1,000
Boost Clock
+62%
2,100
Memory Clock
+44%
2,525
Power Consumption
Recommended PSU
Power Connector
TDP/TGP
-92%
75 W
Rendering
Texture Units (TMU)
+850%
1,216
ROP
16
L2 Cache
8 MB
Benchmarks
MLPerf, llama2-70b-99.9 (Dummy)
MLPerf, llama2-70b-99.9 (fp8)
llama.cpp, llama-2-7b-Q4_0
MLPerf, mixtral-8x7b (fp8)
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
