The AMD Instinct MI300X GPU can handle the Meta LLaMa 60B parameter model on a single GPU. #amd

preview_player
Показать описание

  1. StorageReview
Рекомендации по теме
The AMD Instinct 0:00:25

The AMD Instinct MI300X GPU can handle the Meta LLaMa 60B parameter model on a single GPU. #amd

Introducing AMD Instinct™ 0:01:39

Introducing AMD Instinct™ MI300 Series Accelerators

AMD Reveals MI300X 0:07:25

AMD Reveals MI300X AI Chip (Watch It Here)

AMD's $25,000 GPU: 0:16:24

AMD's $25,000 GPU: Instinct MI210 Tear-Down ft. Level1Techs

Introducing AMD Instinct™ 0:01:02

Introducing AMD Instinct™ MI100 accelerator: First data center GPU to surpass the 10TF FP64 barrier....

Deep-dive into the 0:17:40

Deep-dive into the technology of AMD's MI300

A tour of 0:01:47

A tour of the G383-R80 purpose built for AMD Instinct MI300 APU

AMD vs Nvidia 0:11:40

AMD vs Nvidia -- MI300x vs H100

New GPU for 0:05:32

New GPU for AI Inference & Fine-Tuning LLMs - MI300X

Introducing the AMD 0:01:22

Introducing the AMD Instinct™ MI200 Series Accelerator

AMD's Hidden $100 0:09:23

AMD's Hidden $100 Stable Diffusion Beast!

I'm Buying AMD 0:16:12

I'm Buying AMD Over Nvidia Stock in 2024 (Here's Why)

AMD MI300A: 24 0:05:03

AMD MI300A: 24 Core Zen 4, 128GB HBM3, CDNA3

How this tiny 0:18:00

How this tiny GPU invented the Future

AMD MI300X vs 0:03:12

AMD MI300X vs NVIDIA H100 GPU: Battle for Next Gen AI

AMD's AI Chip 0:08:06

AMD's AI Chip Event: Everything Revealed in 8 Minutes

AMD's Game-Changing AI 0:01:43

AMD's Game-Changing AI GPU: Instinct MI300X with Massive 192GB Memory!

AMD vs. Nvidia: 0:08:41

AMD vs. Nvidia: Battle of the AI Chips

First in Class 0:06:12

First in Class AI GPU? | AMD Instinct MI300X Runs Llama 65B FP16 Inference on Single GPU

AMD INSTINCT MI200 0:01:37

AMD INSTINCT MI200 Unboxing

The Mind Blowing 0:00:41

The Mind Blowing Budget GPU for AI and 3D!

AMD Instinct MI300 0:03:26

AMD Instinct MI300 - OFFICIAL! Unreal POWER!!! CDNA 3 GPU, Up To 24 Zen 4 Cores, 192 GB HBM3 🤯🤯🤯...

MI210s vs A100 0:23:30

MI210s vs A100 -- Is ROCm Finally Viable in 2023? Tested on the Supermicro AS-2114GT-DNR

AMD’s Latest GPUs 0:00:39

AMD’s Latest GPUs are Going to be INSANE for AI and Content Creators

Комментарии
Автор

A single GPU that can beat two of the fastest Nvida H100 GPU's, where MI300X raw computer power is 5, 218 TFlops. This is a Beast!

duladrop
Автор

Interesting form factor. Apparently called OAM(OCP Accelerator Module).

whyjay
Автор

Mi300x has 16 matrix cores per SM compare with only 4 matrix core per SM inside Mi250x.

It has supperior 5 PETA OPS INT8 compare with nvidia h100 &h200 only 4 PETA OPS. Also mi300x has higher fp16 and bfloat 16 compare with competition

kusumayogi
Автор

If it doesn't play Doom, it sucks!

Steven
Автор

Y’all know mofos gonna game this this shit for better looking water

elmoterminator
Автор

At what point are we going to stop calling these GPUs.

TSRB
Автор

is there any possibility to connect the part to a normal mainboard? 🤔

pawelheller
Автор

AMD has taught us not to believe in their performance claims for GPUs, so I won't believe Nvidia isn't going to humiliate this with something that pulls like 0.25x power.

ahmetrefikeryilmaz
Автор

fp16 is 653.7 TFLOPS, H100 nv is 4X that. and lisa can still lie and said they are faster, LOL

MARKXHWANG