Performance Rating
NVIDIA RTX PRO 6000 Blackwell Server outperforms NVIDIA Tesla T4 by 792.39% in the overall GPU ARK performance rating
NVIDIA Tesla T4
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks AdditionalMemory
Memory Size
+500%
96
ГБ
Memory Type
Memory Bandwidth
1.79 TB/s
Memory Bus Width
ML Performance
FP16 (Half Precision)
+93%
126.0
TFLOPS
BF16 (Brain Float)
TF32 (TensorFloat)
Compute Power
FP32 (Single Precision)
+1,448%
126.0
TFLOPS
FP64 (Double Precision)
+674%
1.968
TFLOPS
CUDA Cores
+840%
24,064
RT Cores
+370%
188
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
+370%
188
PCIe Version
ML Software Support
CUDA Version
12.0
Clocks & Performance
Base Clock
+172%
1,590
Boost Clock
+65%
2,617
Memory Clock
+40%
1,750
Power Consumption
TDP/TGP
-88%
70 W
Recommended PSU
-75%
250 W
Power Connector
Rendering
Texture Units (TMU)
+370%
752
ROP
+370%
188
L2 Cache
+3,100%
128 MB
Benchmarks
LLM Perf, Llama-2-7b-hf (bnb)
LLM Perf, Llama-2-7b-hf (gptq)
LLM Perf, Llama-2-7b-hf (unquantized)
LLM Perf, Llama-3.1-8B-Instruct (awq)
LLM Perf, Llama-3.1-8B-Instruct (bnb)
LLM Perf, Llama-3.1-8B-Instruct (gptq)
LLM Perf, Meta-Llama-3-8B (awq)
LLM Perf, Meta-Llama-3-8B (bnb)
LLM Perf, Meta-Llama-3-8B (gptq)
LLM Perf, Meta-Llama-3-8B-Instruct (awq)
LLM Perf, Meta-Llama-3-8B-Instruct (bnb)
LLM Perf, Meta-Llama-3-8B-Instruct (gptq)
MLPerf, llama2-70b-99.9 (fp4)
MLPerf, llama3.1-8b (fp4)
llama.cpp, llama-2-7b-Q4_0
Geekbench AI, FP16
Geekbench AI, INT8
Geekbench AI, FP32
LLM Perf, Mistral-7B-Instruct-v0.2 (gptq)
LLM Perf, Mistral-7B-v0.1 (bnb)
LLM Perf, Mistral-7B-v0.1 (gptq)
LLM Perf, Mistral-7B-v0.1 (unquantized)
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
