正在加载视频...

视频加载失败

返回首页

power consumption at idle with Mac Studio cluster

digitalix's profile picture

Alex Ziskind

22,489 subscribers

623,323 次观看 • 1 年前 •via X (Twitter)

健康养生教育科学技术
LIVE

Anya Rossi• Live Now

Private livecam show

12 条评论

Alex Ziskind 的头像
Alex Ziskind1 年前

Mac Mini Cluster video here (Mac Studio coming soon) M4 Mac Mini CLUSTER 🤯

Pwnage 的头像
Pwnage1 年前

Stormbreaker Max CF design to production Get the next generation of high end performance gaming mice. Shop now:

Anish Gowda 的头像
Anish Gowda1 年前

Hey Alex, Your local LLM content inspired me to build Tiny Ollama Chat - a lightweight UI alternative to OpenWebUI. Only with essentiall features. Open source and actively developing. Would love your thoughts if you get a chance to try it!

Alex Ziskind 的头像
Alex Ziskind1 年前

pretty neat. would be fantastic if you could add some screenshots or video to the repo readme

@EVILeoMORTY@social.tchncs.de 的头像
@[email protected]1 年前

Why is switch using 81W

Alex Ziskind 的头像
Alex Ziskind1 年前

8x 10Gb

moskstraumen 的头像
moskstraumen1 年前

Apple should definitely make its own switches.

Alex Ziskind 的头像
Alex Ziskind1 年前

i know! it would have no ports, but at least will match my silver mac studios 😆

_AG 的头像
_AG1 年前

Can’t you get rid of the network switch and connnect them in thunderbolt5 only ?

Alex Ziskind 的头像
Alex Ziskind1 年前

yes you could

Martin Elitzer 的头像
Martin Elitzer1 年前

Love your content!

Alex Ziskind 的头像
Alex Ziskind1 年前

thx!

相关视频

Llama 3.1 70b beamed to my iPhone from my exo intern home AI cluster of 2 MacBooks and 1 Mac Studio My own private GPT-4 assistant at home / on the go
0:21

Llama 3.1 70b beamed to my iPhone from my exo intern home AI cluster of 2 MacBooks and 1 Mac Studio My own private GPT-4 assistant at home / on the go

Alex Cheema

243,431 次观看 • 1 年前

Starlink Mini $200 Advantages Portability - About the same size as an A4 paper. Low power consumption Idle 20 - 30Watts 40 Watts when running a speed test
0:52

Starlink Mini $200 Advantages Portability - About the same size as an A4 paper. Low power consumption Idle 20 - 30Watts 40 Watts when running a speed test

willard shoko

1,058,000 次观看 • 1 年前

GE90 power-up is insane... sucks in ~1,350 kg of air per second at sea level static/full start, pushing massive airflow even at idle. Sips ~30–60 kg fuel/min total on a 777 at ground idle. This beast is why 777 feels unstoppable. Perfect balance... raw power with efficiency.
0:41

GE90 power-up is insane... sucks in ~1,350 kg of air per second at sea level static/full start, pushing massive airflow even at idle. Sips ~30–60 kg fuel/min total on a 777 at ground idle. This beast is why 777 feels unstoppable. Perfect balance... raw power with efficiency.

Fahad Naim

39,146 次观看 • 4 个月前

G-IDLE LEADER SOYEON COVERED #GDRAGON 's "POWER" AT MBC GAYODAEJUN 2024 . #GDRAGON_POWER
0:37

G-IDLE LEADER SOYEON COVERED #GDRAGON 's "POWER" AT MBC GAYODAEJUN 2024 . #GDRAGON_POWER

GDRAGON'S BUTTERFLY FAN ACCOUNT 🦋(◕ᴗ◕)💛✌

321,155 次观看 • 1 年前

Idle Animation Power Creep #WutheringWaves #Aemeath
1:53

Idle Animation Power Creep #WutheringWaves #Aemeath

Meloo

88,470 次观看 • 4 个月前

HSR Idle Animation Power Creep #honkaistarrail #silverwolf
0:35

HSR Idle Animation Power Creep #honkaistarrail #silverwolf

Meloo

486,219 次观看 • 2 个月前

ชีเสิร์ฟ ๆ CL X NEW MAC STUDIO FIX #MACxCharlotteAustin #ชาล็อตออสติน #FIXผิวใหม่ให้MACเสิร์ฟ
0:21

ชีเสิร์ฟ ๆ CL X NEW MAC STUDIO FIX #MACxCharlotteAustin #ชาล็อตออสติน #FIXผิวใหม่ให้MACเสิร์ฟ

𝗠𝗚𝗜 𝗕𝗲𝘆𝗼𝗻𝗱

64,032 次观看 • 1 年前

DeepSeek-R1 on Mac Studio 192GB 🪄
1:59

DeepSeek-R1 on Mac Studio 192GB 🪄

Georgi Gerganov

259,545 次观看 • 1 年前

The power of Perplexity right at your fingertips, with Visual Studio Code + the Perplexity MCP server in agent mode ✨
2:14

The power of Perplexity right at your fingertips, with Visual Studio Code + the Perplexity MCP server in agent mode ✨

Visual Studio Code

103,484 次观看 • 1 年前

NEWS: Tesla Megapack’s help power the training jobs at xAI’s new 100,000 Nvidia GPU cluster. xAI found millisecond power fluctuations when GPUs start training, causing issues with power infrastructure. The solution was to use generators to charge the Tesla Megapacks, which then discharge to power the training jobs.
0:37

NEWS: Tesla Megapack’s help power the training jobs at xAI’s new 100,000 Nvidia GPU cluster. xAI found millisecond power fluctuations when GPUs start training, causing issues with power infrastructure. The solution was to use generators to charge the Tesla Megapacks, which then discharge to power the training jobs.

Sawyer Merritt

1,084,876 次观看 • 1 年前

Looking at some friends with Mac
0:33

Looking at some friends with Mac

dr. coffee ☕

18,665 次观看 • 2 年前

ผู้นำเทรน555555555 CL X NEW MAC STUDIO FIX #MACxCharlotteAustin
0:21

ผู้นำเทรน555555555 CL X NEW MAC STUDIO FIX #MACxCharlotteAustin

alif /✿ ‎੯‧̀͡ ζྀི)

57,201 次观看 • 1 年前

iPad mini + Mac Studio = Macintosh [📹 bbbigdeer]
0:53

iPad mini + Mac Studio = Macintosh [📹 bbbigdeer]

Massimo

12,332,110 次观看 • 2 年前

M4 Mac AI Coding Cluster Uses EXO Labs to run LLMs (here Qwen 2.5 Coder 32B at 18 tok/sec) distributed across 4 M4 Mac Minis (Thunderbolt 5 80Gbps) and a MacBook Pro M4 Max. Local alternative to Cursor (benchmark comparison soon).
0:20

M4 Mac AI Coding Cluster Uses EXO Labs to run LLMs (here Qwen 2.5 Coder 32B at 18 tok/sec) distributed across 4 M4 Mac Minis (Thunderbolt 5 80Gbps) and a MacBook Pro M4 Max. Local alternative to Cursor (benchmark comparison soon).

Alex Cheema

517,325 次观看 • 1 年前

Wake the world's sleeping compute. Look at the Mac nearest to you. What's it doing? Probably nothing. There are 100M+ Macs with Apple Silicon out there. Apple quietly made them *really* good at inference. A $3k Mac runs a 60B model at 30 watts. Most sit idle most of the day. Meanwhile every AI API call passes through three layers of margin before reaching the hardware. We call this the Inference Tax. We got curious: what happens if you connect idle Macs directly to inference demand? This is Darkbloom. Private inference network for idle Macs. darkbloom [dot] dev -- paper + code open. Reply for invite + free credits ↓
2:58

Wake the world's sleeping compute. Look at the Mac nearest to you. What's it doing? Probably nothing. There are 100M+ Macs with Apple Silicon out there. Apple quietly made them *really* good at inference. A $3k Mac runs a 60B model at 30 watts. Most sit idle most of the day. Meanwhile every AI API call passes through three layers of margin before reaching the hardware. We call this the Inference Tax. We got curious: what happens if you connect idle Macs directly to inference demand? This is Darkbloom. Private inference network for idle Macs. darkbloom [dot] dev -- paper + code open. Reply for invite + free credits ↓

Gajesh

505,721 次观看 • 2 个月前

🚨THIS IS REALLY HEATING UP -Tehran's power knocked out in large sections -CLUSTER BOMBS filled with "bomblets' pummeling Israel.
1:00

🚨THIS IS REALLY HEATING UP -Tehran's power knocked out in large sections -CLUSTER BOMBS filled with "bomblets' pummeling Israel.

Tablesalt 🇨🇦🇺🇸

29,310 次观看 • 3 个月前

PJ is having a rough time with Cluster in the studio... Hopefully he can take it a bit more seriously next time!! #Fortntie #FortniteArt
0:22

PJ is having a rough time with Cluster in the studio... Hopefully he can take it a bit more seriously next time!! #Fortntie #FortniteArt

🐟Cozy Muds🐰 Comms OPEN

42,358 次观看 • 14 天前

Boldy James and Mac Miller talking about this studio session
1:56

Boldy James and Mac Miller talking about this studio session

Steez⁴⁷

94,425 次观看 • 1 年前

New M3 Ultra Mac Studio is 🔥🔥
1:01

New M3 Ultra Mac Studio is 🔥🔥

iJustine

59,420 次观看 • 1 年前

just generated an ai video on my mac no cloud, no server bill last year this needed a gpu cluster
0:36

just generated an ai video on my mac no cloud, no server bill last year this needed a gpu cluster

Igor Tarasenko

426,394 次观看 • 1 年前