AMD Radeon Instinct MI250 vs GIGABYTE AORUS RTX 3060 ELITE

Comparison of AMD Radeon Instinct MI250 with 128 GB HBM2e and 13,312 cores vs GIGABYTE AORUS RTX 3060 ELITE with 12 GB GDDR6 and 3,584 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

AMD Radeon Instinct MI250

50.9

AMD Radeon Instinct MI250

50.9
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

GIGABYTE AORUS RTX 3060 ELITE

GIGABYTE AORUS RTX 3060 ELITE

Contents:

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

Memory

Memory Size

🔥 +967% 128 ГБ
12 GB

Memory Type

HBM2e GDDR6

Memory Bandwidth

🔥 3.28 TB/s
360.0 GB/s

Memory Bus Width

8,192 бит 192 бит

ML Performance

FP16 (Half Precision)

🔥 +2,606% 362.1 TFLOPS
13.38 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +238% 45.26 TFLOPS
13.38 TFLOPS

FP64 (Double Precision)

🔥 +21,545% 45.26 TFLOPS
0.2091 TFLOPS

CUDA Cores

🔥 +271% 13,312
3,584

RT Cores

No 28

Architecture & Compatibility

GPU Architecture

CDNA 2.0 Ampere

SM (Streaming Multiprocessor)

No 28

PCIe Version

PCIe 4.0 x16 PCIe 4.0 x16

ML Software Support

CUDA Version

No 8.6

Clocks & Performance

Base Clock

🔥 1,000
1,320

Boost Clock

🔥 1,700
1,867

Memory Clock

🔥 1,600
1,875

Power Consumption

TDP/TGP

500 W
🔥 -66% 170 W

Recommended PSU

900 W
🔥 -50% 450 W

Power Connector

2x 8-pin 1x 6-pin + 1x 8-pin

Rendering

Texture Units (TMU)

🔥 +643% 832
112

ROP

No 28

L2 Cache

🔥 +433% 16 MB
3 MB

Benchmarks

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

63.9 tokens/s

Additional

Slots

Dual-slot Triple-slot

Release Date

Nov. 8, 2021 Jan. 12, 2021

Display Outputs

No outputs
1x HDMI 2.1
3x DisplayPort 1.4a

2x HDMI 2.1
2x DisplayPort 1.4a

Renting is cheaper than buying