latchkey

19 ms·

361.
by latchkey 8mo ago
maybe i lean too much in one direction, but what is a "legitimate tax"? Once again, count on hn for the downvotes. Yep, those shall not speak of downvotes, or taxation.
362.
by latchkey 8mo ago
https://oj-hn.com
363.
by latchkey 8mo ago
It is on my feature list for https://oj-hn.com as well.
364.
by latchkey 8mo ago
This is super cool. In other similar news, I've been working on enhancing the HN ux, but still in the browser as an extension. The current build up on the Chrome store is pretty stable. https://oj-hn.com
365.
by latchkey 8mo ago
@dang at least the RS story vs. paywall please.
366.
by latchkey 8mo ago
command-space... type "keychain access"
367.
by latchkey 8mo ago
I can give you access to MI300x. Also, reach out to Anush please. He commented on here.
368.
by latchkey 8mo ago
Nice work! Love it.
369.
by latchkey 8mo ago
Like Uber/Airbnb in early days, this is heavily subsidized.
370.
by latchkey 8mo ago
Anush is who you want to ping, he's motivated and will take care of you. https://x.com/AnushElangovan
371.
by latchkey 8mo ago
biome kind of does this stuff for me in the typescript world. for example, it recommends for...of instead of forEach. paired with ultracite, it really improves my workflow because it is super easy to add to any project, just one dependency.
372.
by latchkey 8mo ago
Note that this targets GFX11, which is RDNA3. Great for consumer, but not the enterprise (CDNA) level at all. In other words, not a "cuda moat killer".
373.
by latchkey 8mo ago
See also: https://scale-lang.com/ Write CUDA code. Run Everywhere. Your CUDA skills are now universal. SCALE compiles your unmodified applications to run natively on any accelerator, ending the nightmare of maintaining mult
374.
Speed Is the Moat: Inference Performance on AMD GPUs(amd.com)
3 points by latchkey 8mo ago|0 comments
375.
by latchkey 8mo ago
Ok, tell me the value.
376.
Unitree robot's martial arts performance at the Chinese New Year gala(twitter.com)
2 points by latchkey 8mo ago|0 comments
377.
by latchkey 8mo ago
It has updates at the bottom. Most recently 14 Sep 2025.
378.
Former Karaoke Company Drags Logistics into the 'AI Scare Trade'(finance.yahoo.com)
1 points by latchkey 8mo ago|0 comments
379.
by latchkey 8mo ago
I do enjoy this, but the title is such clickbait. I was running websites on a sparc 2 back in 1995.
380.
by latchkey 8mo ago
Even the 3rd party AI benchmarks that are published [0], are all sham too. It is run by a paid shill (semianalysis) and all highly tuned by the vendors to make themselves look good. [0] https://github.com/InferenceMAX/I
381.
by latchkey 8mo ago
Neat, but you can just do this in mermaid too. Taking one of your examples: <mermaid> flowchart LR web([Frontend]) subgraph platform [Cloud Platform] api([API Server]) db[(Database)] api --> db end web
382.
by latchkey 8mo ago
Oh come on, what is this. Affect my mindset how exactly?
383.
by latchkey 8mo ago
I remembered $1m from when I was in their booth at SC24, but when I just looked, I was wrong. It is worse... https://www.datacenterdynamics.com/en/news/cerebras-unveils-...
384.
by latchkey 8mo ago
Exactly my point. Their architecture requires someone to invest the capex / opex to also build another hall.
385.
by latchkey 8mo ago
AMD
386.
by latchkey 8mo ago
TPUs don't have enough memory either, but they have really great interconnects, so they can build a nice high density cluster. Compare the photos of a Cerebras deployment to a TPU deployment. https://www.nextplatform.com&#x2
387.
by latchkey 8mo ago
The dirty secret is that there is plenty of power. But, it isn't all in one place and it is often stranded in DC's that can't do the density needed for AI compute. Training models needs everything in one DC, inference doesn&#
388.
by latchkey 8mo ago
Exactly. They won't ever tell you. It is never published. Let's not forget that the CEO is an SEC felon who got caught trying to pull a fast one.
389.
by latchkey 8mo ago
Not for what they are using it for. It is $1m+/chip and they can fit 1 of them in a rack. Rack space in DC's is a premium asset. The density isn't there. AI models need tons of memory (this product annoucement is case in poin
390.
by latchkey 8mo ago
Maybe, but he also drinks from the firehose.

More ›