AMD Radeon Instinct MI250 vs NVIDIA H100 CNX

Comparison of AMD Radeon Instinct MI250 with 128 GB HBM2e and 13,312 cores vs NVIDIA H100 CNX with 80 GB HBM2e and 14,592 cores.

Loading...

Performance Rating

AMD Radeon Instinct MI250 outperforms NVIDIA H100 CNX by 12.41% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Instinct MI250

50.9

AMD Radeon Instinct MI250

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

NVIDIA H100 CNX

45.3

NVIDIA H100 CNX

45.3

Contents:

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

Memory

Memory Size

🔥 +60% 128 ГБ
80 ГБ

Memory Type

HBM2e HBM2e

Memory Bandwidth

🔥 +61% 3.28 TB/s
2.04 TB/s

Memory Bus Width

8,192 бит 5,120 бит

ML Performance

FP16 (Half Precision)

🔥 +68% 362.1 TFLOPS
215.4 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

45.26 TFLOPS
🔥 +19% 53.84 TFLOPS

FP64 (Double Precision)

🔥 +68% 45.26 TFLOPS
26.92 TFLOPS

CUDA Cores

13,312
🔥 +10% 14,592

RT Cores

No No

Architecture & Compatibility

GPU Architecture

CDNA 2.0 Hopper

SM (Streaming Multiprocessor)

No
🔥 114

PCIe Version

PCIe 4.0 x16 PCIe 5.0 x16

ML Software Support

CUDA Version

No 9.0

Clocks & Performance

Base Clock

🔥 +45% 1,000
690

Boost Clock

1,700
🔥 +9% 1,845

Memory Clock

🔥 1,600
1,593

Power Consumption

TDP/TGP

500 W
🔥 -30% 350 W

Recommended PSU

900 W
🔥 -17% 750 W

Power Connector

2x 8-pin 8-pin EPS

Rendering

Texture Units (TMU)

🔥 +82% 832
456

ROP

No No

L2 Cache

16 MB
🔥 +212% 50 MB

Benchmarks

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

63.9 tokens/s

Additional

Slots

Dual-slot Dual-slot

Release Date

Nov. 8, 2021 March 21, 2023

Display Outputs

No outputs
No outputs

Renting is cheaper than buying