NVIDIA RTX PRO 6000 Blackwell Server vs NVIDIA T400

Comparison of NVIDIA RTX PRO 6000 Blackwell Server with 96 GB GDDR7 and 24,064 cores vs NVIDIA T400 with 2 GB GDDR6 and 384 cores.

Loading...

Performance Rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA RTX PRO 6000 Blackwell Server

57.5

NVIDIA RTX PRO 6000 Blackwell Server

57.5
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA T400

0.0

NVIDIA T400

0.0

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Benchmarks Additional

Memory

Memory Size

🔥 +4,700% 96 ГБ
2 ГБ

Memory Type

GDDR7 GDDR6

Memory Bandwidth

🔥 1.79 TB/s
80.00 GB/s

Memory Bus Width

512 бит 64 бит

ML Performance

FP16 (Half Precision)

🔥 +5,656% 126.0 TFLOPS
2.189 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 126.0 TFLOPS
No

FP64 (Double Precision)

🔥 +5,654% 1.968 TFLOPS
0.0342 TFLOPS

CUDA Cores

🔥 +6,167% 24,064
384

RT Cores

🔥 188
No

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Turing

SM (Streaming Multiprocessor)

🔥 +3,033% 188
6

PCIe Version

PCIe 5.0 x16 PCIe 3.0 x16

ML Software Support

CUDA Version

🔥 12.0
7.5

Clocks & Performance

Base Clock

🔥 +279% 1,590
420

Boost Clock

🔥 +84% 2,617
1,425

Memory Clock

🔥 +40% 1,750
1,250

Power Consumption

TDP/TGP

600 W
🔥 -95% 30 W

Recommended PSU

1000 W
🔥 -80% 200 W

Power Connector

1x 16-pin None

Rendering

Texture Units (TMU)

🔥 +3,033% 752
24

ROP

🔥 188
No

L2 Cache

🔥 128 MB
1024 KB

Benchmarks

MLPerf, llama2-70b-99.9 (fp4)

3 250 tokens/s

MLPerf, llama3.1-8b (fp4)

5 758 tokens/s

Geekbench AI, FP16

53 322 points

Geekbench AI, INT8

28 264 points

Geekbench AI, FP32

37 299 points

MLPerf, mixtral-8x7b (fp8)

3 767 tokens/s

Additional

Slots

Dual-slot Single-slot

Release Date

March 18, 2025 May 6, 2021

Display Outputs

4x DisplayPort 2.1b
3x mini-DisplayPort 1.4a

Renting is cheaper than buying