AMD Radeon Instinct MI300X vs NVIDIA Quadro DCC

Comparison AMD Radeon Instinct MI300X with 192 GB HBM3 and 19,456 cores vs NVIDIA Quadro DCC with 64 GB SDR.

Loading...

Performance Rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Instinct MI300X

94.1

AMD Radeon Instinct MI300X

94.1
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA Quadro DCC

NVIDIA Quadro DCC

Expert Comparison

AMD Radeon Instinct MI300X обладает значительно большей пропускной способностью и больше памяти (192 ГБ HBM3 против 64 ГБ SDR), что делает его предпочтительным для вычислений и научных вычислений. Его архитектура CDNA 3.0 также более современная, чем Kelvin у NVIDIA Quadro DCC. Однако Quadro DCC более старый и менее энергоэффективный, с неизвестным TDP и устаревшей шиной 128 бит.

Contents:

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

Memory

Memory Size

🔥 +200% 192 ГБ
64 MB

Memory Type

HBM3 SDR

Memory Bandwidth

🔥 +180% 10.3 TB/s
3.680 GB/s

Memory Bus Width

8,192 бит 128 бит

ML Performance

FP16 (Half Precision)

🔥 653.7 TFLOPS
No

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 81.72 TFLOPS
No

FP64 (Double Precision)

🔥 81.72 TFLOPS
No

CUDA Cores

🔥 19,456
No

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 3.0 Kelvin

SM (Streaming Multiprocessor)

No No

PCIe Version

PCIe 5.0 x16 AGP 4x

ML Software Support

CUDA Version

No No

CUDA Toolkit status

Supported Supported

Clocks & Performance

Base Clock

🔥 1,000
No

Boost Clock

🔥 2,100
No

Memory Clock

🔥 +998% 2,525
230

Power Consumption

Recommended PSU

1150 W
🔥 -83% 200 W

Power Connector

None None

TDP/TGP

750 W unknown

Rendering

Texture Units (TMU)

🔥 +30,300% 1,216
4

ROP

No No

L2 Cache

🔥 16 MB
No

Benchmarks

MLPerf, llama2-70b-99.9 (UNSET)

1 983 tokens/s

MLPerf, llama2-70b-99.9 (fp16)

1 740 tokens/s

MLPerf, llama2-70b-99.9 (fp8)

1 057 tokens/s

MLPerf, llama3.1-405b (UNSET)

30.4 tokens/s

MLPerf, llama3.1-405b (fp16)

34.8 tokens/s

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

232.9 tokens/s

MLPerf, mixtral-8x7b (fp8)

5 975 tokens/s

Additional

Slots

OAM Module Single-slot

Release Date

Dec. 6, 2023 May 14, 2001

Display Outputs

No outputs
1x DVI
1x VGA
1x S-Video

Renting is cheaper than buying