AMD Radeon Instinct MI325X vs Sapphire NITRO+ RX 6950 XT PURE

Comparison of AMD Radeon Instinct MI325X with 288 GB HBM3e and 19,456 cores vs Sapphire NITRO+ RX 6950 XT PURE with 16 GB GDDR6 and 5,120 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI325X

AMD Radeon Instinct MI325X

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

Sapphire NITRO+ RX 6950 XT PURE

Sapphire NITRO+ RX 6950 XT PURE

Contents:

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

Memory

Memory Size

288 GB 16 GB

Memory Type

HBM3e GDDR6

Memory Bandwidth

10.3 TB/s 576.0 GB/s

Memory Bus Width

8,192 бит 256 бит

ML Performance

FP16 (Half Precision)

653.7 TFLOPS 48.5 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

81.72 TFLOPS 24.25 TFLOPS

FP64 (Double Precision)

81.72 TFLOPS 1.516 TFLOPS

CUDA Cores

19,456 5,120

RT Cores

No 80

Architecture & Compatibility

GPU Architecture

CDNA 3.0 RDNA 2.0

SM (Streaming Multiprocessor)

No No

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x16

ML Software Support

CUDA Version

No No

Clocks & Performance

Base Clock

1,000 2,009

Boost Clock

2,100 2,368

Memory Clock

2,525 2,250

Power Consumption

TDP/TGP

1000 W
🔥 -66% 335 W

Recommended PSU

1400 W
🔥 -50% 700 W

Power Connector

None 1x 6-pin + 2x 8-pin

Rendering

Texture Units (TMU)

1,216 320

ROP

No 80

L2 Cache

16 MB 4 MB

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 Quad-slot

Release Date

Oct. 12, 2024 May 10, 2022

Display Outputs

No outputs
1x HDMI 2.1
2x DisplayPort 1.4a

Renting is cheaper than buying