Performance Rating
The H200 NVL is a professional graphics card by NVIDIA, launched on November 18th, 2024. Built on the 5 nm process, and based on the GH100 graphics processor, the card does not support DirectX. Since H200 NVL does not support DirectX 11 or DirectX 12, it might not be able to run all the latest games. The GH100 graphics processor is a large chip with a die area of 814 mm² and 80,000 million transistors. It features 16896 shading units, 528 texture mapping units, and 24 ROPs. Also included are 528 tensor cores which help improve the speed of machine learning applications. NVIDIA has paired 141 GB HBM3e memory with the H200 NVL, which are connected using a 6144-bit memory interface. The GPU is operating at a frequency of 1365 MHz, which can be boosted up to 1785 MHz, memory is running at 1593 MHz.
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks AdditionalMemory
Memory Size
Memory Type
Memory Bandwidth
Memory Bus Width
ML Performance
FP16 (Half Precision)
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
FP64 (Double Precision)
CUDA Cores
RT Cores
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
PCIe Version
ML Software Support
CUDA Version
Clocks & Performance
Base Clock
Boost Clock
Memory Clock
Power Consumption
TDP/TGP
Recommended PSU
Power Connector
Benchmarks
MLPerf, llama2-70b-99.9 (fp4)
MLPerf, llama3.1-405b (fp4)
MLPerf, deepseek-r1 (fp4)
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
