Did you catch that?
The new maxed out 256GB M5 Ultra Studio can be leased for *basically* the same price as a Max subscription across providers.
For $257 a month, you can have on device intelligence that never runs out of tokens, and sips power.
Best part is you can buy out the lease at the end of the lease for a $0 penalty. That's right 0% interest.
So if the price is $12.9K and your lease is $9.9K over 36 months, then you pay the $3K difference and you now own that device.
Or you can trade up into the M8 Ultra or whatever.
I'm excited for this. 1.2 TB/s is pretty impressive. Can't wait to see some real benchmarks and testing on this.
I'm waiting for the 512GB.
we didn't know how to design a chip a couple months ago
so we reinvented Groq's LPU to run inference on @karpathy 's MicroGPT!
here's a breakdown of what we learned:
@samanthaaouyang wait this is actually wild. I've also gotten my fair share of dms on linkedin where the person clearly has done ZERO due diligence ๐คฃ
@chrislevan24 Applied! Curious whether new signal types usually come from reps asking for something specific after a lost deal, or from the data side spotting a pattern first
@hunterchen_7 Yeah the latter is obviously harder to quantify but when everyone is worried about tok/sec and throughput I feel like this is something that might become a bigger deal?
Just due to the atomic nature of what people are using to measure progress.
Wait correct me if I'm wrong, but isn't time to first token a flawed metric for chat bots?
Like I understand that the whole point is to measure when the first token comes out, but what if the first few tokens of output are just filler..
Is there another term for this? ๐
My favourite thing to do after I get obliterated by the biweekly leetcode contest is to go through the top 10 contestants and watch how they don't even try to hide that they're just copy and pasting from claude ๐คฃ๐คฃ
@LukeBlomm hmm taking away a part of the graph and seeing how it reacts. I've seen this somewhere before, perhaps on a really tuff paper about roster topology
@iveybusiness the intro animation I built in LottieLab. it's all live at https://t.co/D9mUKntFnx now aligned with Morrissette. Thought I'd share since I'm taking a backseat to a new set of developers that will refine what we started!
@iveybusiness So I rebuilt it. the wireframe look came from the schematic drawings in the tunnel between the Amit Chakma & Spencer eng buildings, creative but institutional. the loading animation + spinning building were inspired by the DeerHacks 2024 site https://t.co/2ODQdcISq5