AMD Radeon Instinct MI300X vs EVGA GTX 570 HD Superclocked

Comparison of AMD Radeon Instinct MI300X with 192 GB HBM3 and 19,456 cores vs EVGA GTX 570 HD Superclocked with 1,280 GB GDDR5 and 480 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI300X

AMD Radeon Instinct MI300X

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

EVGA GTX 570 HD Superclocked

EVGA GTX 570 HD Superclocked

Contents:

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

Memory

Memory Size

192 GB 1,280 MB

Memory Type

HBM3 GDDR5

Memory Bandwidth

10.3 TB/s 156.0 GB/s

Memory Bus Width

8,192 бит 320 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 1.53 TFLOPS

FP64 (Double Precision)

81.72 TFLOPS 0.1913 TFLOPS

CUDA Cores

19,456 480

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 3.0 Fermi 2.0

SM (Streaming Multiprocessor)

No 15

PCIe Version

PCIe 5.0 x16 PCIe 2.0 x16

ML Software Support

CUDA Version

No 2.0

Clocks & Performance

Base Clock

1,000 No

Boost Clock

2,100 No

Memory Clock

2,525 975

Power Consumption

TDP/TGP

750 W
🔥 -71% 219 W

Recommended PSU

1150 W
🔥 -52% 550 W

Power Connector

None 2x 6-pin

Rendering

Texture Units (TMU)

1,216 60

ROP

No No

L2 Cache

16 MB 640 KB

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 Dual-slot

Release Date

Dec. 6, 2023 Feb. 15, 2011

Display Outputs

No outputs
2x DVI
1x HDMI 1.3a
1x DisplayPort

Renting is cheaper than buying