Performance Rating
NVIDIA H200 SXM 141 GB outperforms NVIDIA Tesla M40 by 1,080.04% in the overall GPU ARK performance rating
NVIDIA Tesla M40
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks AdditionalMemory
Memory Size
+1,075%
141
ГБ
Memory Type
Memory Bandwidth
4.89 TB/s
Memory Bus Width
ML Performance
FP16 (Half Precision)
267.6
TFLOPS
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
+879%
66.91
TFLOPS
FP64 (Double Precision)
+15,567%
33.45
TFLOPS
CUDA Cores
+450%
16,896
RT Cores
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
132
PCIe Version
ML Software Support
CUDA Version
9.0
Clocks & Performance
Base Clock
+58%
1,500
Boost Clock
+78%
1,980
Memory Clock
+6%
1,593
Power Consumption
TDP/TGP
-64%
250 W
Recommended PSU
-45%
600 W
Power Connector
Rendering
Texture Units (TMU)
+175%
528
ROP
L2 Cache
+1,567%
50 MB
Benchmarks
MLPerf, llama2-70b-99.9 (UNSET)
MLPerf, llama2-70b-99.9 (fp16)
MLPerf, llama2-70b-99.9 (fp8)
MLPerf, llama3.1-405b (fp16)
MLPerf, llama3.1-405b (fp8)
MLPerf, llama3.1-8b (fp8)
llama.cpp, gpt-oss 20B Q4_K - Medium
llama.cpp, llama 7B Q4_0
llama.cpp, llama-2-7b-Q4_0
llama.cpp, qwen3 32B Q4_K - Medium
llama.cpp, qwen3moe 30B.A3B Q4_K - Medium
MLPerf, deepseek-r1 (fp8)
MLPerf, mixtral-8x7b (fp8)
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
