Exo Labs
@exolabs_
Followers
8,745
Following
3
Media
4
Statuses
59
Run your own AI cluster at home using everyday devices
Joined March 2024
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
バニーの日
• 227096 Tweets
STEP DOWN FASCIST HASINA
• 80108 Tweets
ブラジル
• 65280 Tweets
Hayırlı Cumalar
• 56856 Tweets
ヒロアカ
• 56284 Tweets
#يوم_الجمعه
• 38669 Tweets
パンツの日
• 35225 Tweets
#erişim
• 34615 Tweets
#instagramdown
• 31026 Tweets
長岡花火
• 29403 Tweets
デュエマ
• 21200 Tweets
騎乗停止
• 18633 Tweets
芝コース
• 18588 Tweets
角田大河
• 18247 Tweets
日本バスケ
• 16279 Tweets
男子バスケ
• 15529 Tweets
Teddy Riner
• 14444 Tweets
フランス戦
• 13464 Tweets
Osmar Olvera
• 13069 Tweets
逆バニー
• 12988 Tweets
飲酒運転
• 11754 Tweets
Llama-3-405b is coming on 23rd July and exo will support it on day 0.
You don't need a H100 to run Llama-3-405b.
2 MacBooks and 1 Mac Studio will do the job, with
@exolabs_
to aggregate the memory/compute.
I'm ready for you, Llama-3-405b.
41
114
1K
4
18
79
New exo feature just dropped.
@exolabs_
now tracks in real-time if you are GPU poor or GPU rich based on all the devices connected in your AI cluster.
Here I have 2 MacBook Pro’s, 1 MacBook Air and 1 Max Studio connected.
h/t
@caseykcaruso
and
@huggingface
for inspiring this.
5
11
70
2
4
45
Mixture of Expert (MOE) models + distributed inference = a match made in heaven.
Soon, you’ll be able to run 100b+ parameter models like this on normal laptops / phones with exo.
Track the GitHub issue here:
@ac_crypto
@exolabs_
It's an MOE so only 21B active params. Actually an interesting candidate for distributed inference. Easier to make it faster sharding across experts
1
0
9
1
4
37
Does anyone have 8 maxed out mac studios?
@BasedBeffJezos
wants to know what we could do with them.
We can aggregate the memory/compute on exo and they would have almost as much compute and 20x the memory of a H100.
5
4
40
From clean MacBooks to chat interface in 60 seconds
How long does it take to get distributed inference running locally across 2 MacBook GPUs from a fresh install?
About 60 seconds, running
@exolabs_
Watch till the end, I chat to the cluster using
@__tinygrad__
ChatGPT web interface
Code is open source 👇
17
51
627
1
4
33
Exo was featured in Tom’s Hardware.
“Thanks to the work of a team of developers, a new software could allow you to run your own AI cluster at home using your existing smartphones, tablets, and computers.”
Link to repo:
New software lets you run a private AI cluster at home with networked smartphones, tablets, and computers — Exo software runs LLama and other AI models
0
3
17
1
4
29
@ac_crypto
The new live network topology view also shows which device is active in the inference in real-time.
3
0
12
@ac_crypto
@__tinygrad__
Got to try this with 100b+ parameter Mixture of Expert (MOE) models next h/t
@awnihannun
Mixture of Expert (MOE) models + distributed inference = a match made in heaven.
Soon, you’ll be able to run 100b+ parameter models like this on normal laptops / phones with exo.
Track the GitHub issue here:
1
4
37
1
0
5
