Performance Rating
NVIDIA RTX PRO 4000 Blackwell SFF outperforms Gainward RTX 3060 Ti Chasing Wind G by 46.47% in the overall GPU ARK performance rating
NVIDIA RTX PRO 4000 Blackwell SFF
The NVIDIA RTX PRO 4000 Blackwell SFF significantly outperforms the Gainward RTX 3060 Ti Chasing Wind G, leading by 46% in the overall GPI performance index. For machine learning, the NVIDIA RTX PRO 4000 Blackwell SFF has the edge in VRAM (24 vs 8 GB) and tensor cores (280 vs 152). For gaming, the NVIDIA RTX PRO 4000 Blackwell SFF delivers 58% more FP32 throughput, and the NVIDIA RTX PRO 4000 Blackwell SFF has 84% more RT cores. In performance per watt, the NVIDIA RTX PRO 4000 Blackwell SFF is 353% more efficient at 70W TDP vs 200W for the Gainward RTX 3060 Ti Chasing Wind G.
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering AdditionalMemory
Memory Size
+200%
24
ГБ
Memory Type
Memory Bandwidth
Memory Bus Width
ML Performance
FP16 (Half Precision)
+58%
25.66
TFLOPS
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
+58%
25.66
TFLOPS
FP64 (Double Precision)
+58%
0.401
TFLOPS
CUDA Cores
+84%
8,960
RT Cores
+84%
70
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
+84%
70
PCIe Version
ML Software Support
CUDA Version
12.0
CUDA Toolkit (first supported)
CUDA Toolkit status
Clocks & Performance
Base Clock
+144%
1,410
Boost Clock
+16%
1,665
Memory Clock
+56%
1,750
Power Consumption
Recommended PSU
-55%
250 W
Power Connector
TDP/TGP
-65%
70 W
Rendering
Texture Units (TMU)
+84%
280
ROP
+84%
70
L2 Cache
+1,100%
48 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
