AMD Radeon Instinct MI325X vs Mercury GSC6201 6U OpenVPX

Comparison of AMD Radeon Instinct MI325X with 288 GB HBM3e and 19,456 cores vs Mercury GSC6201 6U OpenVPX with 2 GB GDDR5 and 240 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI325X

100.0

AMD Radeon Instinct MI325X

100.0
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

Mercury GSC6201 6U OpenVPX

Mercury GSC6201 6U OpenVPX

Contents:

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

Memory

Memory Size

🔥 +14,300% 288 ГБ
2 GB ×2 (4 ГБ)

Memory Type

HBM3e GDDR5

Memory Bandwidth

🔥 10.3 TB/s
80.00 GB/s ×2 (160 GB/s)

Memory Bus Width

8,192 бит 256 бит ×2 (512 бит)

ML Performance

FP16 (Half Precision)

🔥 653.7 TFLOPS
No

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +18,817% 81.72 TFLOPS
0.432 TFLOPS

FP64 (Double Precision)

🔥 +226,900% 81.72 TFLOPS
0.036 TFLOPS

CUDA Cores

🔥 +8,007% 19,456
240 ×2 (480)

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 3.0 Fermi

SM (Streaming Multiprocessor)

No 5

PCIe Version

PCIe 5.0 x16 MXM-B (3.0)

ML Software Support

CUDA Version

No 2.1

Clocks & Performance

Base Clock

🔥 1,000
No

Boost Clock

🔥 2,100
No

Memory Clock

🔥 +304% 2,525
625

Power Consumption

TDP/TGP

1000 W
🔥 -85% 150 W

Recommended PSU

1400 W No

Power Connector

None None

Rendering

Texture Units (TMU)

🔥 +2,940% 1,216
40 ×2 (80)

ROP

No No

L2 Cache

🔥 16 MB
512 KB

Benchmarks

MLPerf, llama2-70b-99.9 (Dummy)

3 596 tokens/s

MLPerf, llama2-70b-99.9 (fp8)

1 946 tokens/s

llama.cpp, llama-2-7b-Q4_0

22.4 tokens/s

MLPerf, mixtral-8x7b (fp8)

6 975 tokens/s

Additional

Slots

OAM Module MXM Module

Release Date

Oct. 15, 2024 Feb. 22, 2011

Display Outputs

No outputs
Portable Device Dependent

Renting is cheaper than buying