LarsDu88

5 ms·

61.
by LarsDu88 2mo ago
The hyperscalers already had the compute monopoly. They just spent a bunch of money on even more compute. Compute is OK because it's reasonably general purpose to reallocate for what comes after chatbots (e.g. consumer robotics which i
62.
by LarsDu88 2mo ago
This is a bit of a doomer article, but quite honestly, 200 billion dollars a year on a 30 billion dollar a year business that is growing does not really sound as bad as the author makes it out to be, especially when that business consists o
63.
by LarsDu88 2mo ago
Wow screw lightbulb is lower than tie bag or close ziploc. Pretty surprising.
64.
by LarsDu88 2mo ago
Where did you get this perf metric? I couldn't find it in the release
65.
by LarsDu88 2mo ago
All VLM, VLA models. I wonder if these architectures will reach prod before Yann LeCunn's world model JEPA startup even gets off the ground in Europe.
66.
by LarsDu88 3mo ago
Certainly there will be demand for Fable7, but that demand is context specific. Frontier labs' profit is dependent on there being sufficient demand for the next layer of capability and whether the premium consumers are willing to pay f
67.
by LarsDu88 3mo ago
At 9000 tokens/s you could interleave a lot of requests so long as pre-fill is also fast. It really depends on how much you need to keep sessions open to take advantage of KV caching
68.
by LarsDu88 3mo ago
Currently the setup is paged view in RAM shuttled to HBDRAM (VRAM) on the GPU, which in turn has to get materialized piece by piece onto cache SRAM on the GPU. Cerebras tries to get around this by keeping everything on cache SRAM as much as
69.
by LarsDu88 3mo ago
Cerebras is not an ASIC. It is a large wafer scale chip that has a load of small SRAM modules paired with tiny compute models. The SRAM is basically a giant cache that is supposed to eliminate the bottleneck between shuttling and materializ
70.
by LarsDu88 3mo ago
Honestly, whether you think burning current SOTA to hardware is an overinvestment risk depends on what your definition of intelligence is. If you think intelligence is something that can grow like height such that 18 months from now we will
71.
by LarsDu88 3mo ago
The closest example I've seen is ChatJimmy: https://chatjimmy.ai/ a prototype from Taalas running Llama 8B Scaling this up to 2.8 Trillion (350X increase), will certainly be challenging. If I was younger and had the ri
72.
by LarsDu88 3mo ago
They commoditized their compliment, but made sure their chips only worked with their accessory products, particularly in the 70s
73.
by LarsDu88 3mo ago
cough Ub Iwerks
74.
by LarsDu88 3mo ago
The open weight, open architecture releases of the past several days has me more convinced that ultimately, the winner will be whoever burns their models to ASICs fastest. The LLMs themselves are capable of doing some aspects of chip design
75.
by LarsDu88 3mo ago
You mean Jack Kirby right? LoL
76.
by LarsDu88 3mo ago
I don't get this "build a platform" nonsense. He's a designer. A good one. Who ripped off much of his aethetics from vintage Braun products.
77.
by LarsDu88 3mo ago
Andry Grove was great, but Intel really was the epitome of "competition is for losers" In the early days all their products were explicitly designed to only work with each other to create a hardware walled garden.
78.
by LarsDu88 3mo ago
So build the full model with SRAM and all?
79.
by LarsDu88 3mo ago
I believes the weights are burned as ROM microcode, but for an effective inference speedup, you do want to burn the architecture (matmuls, activation functions, MoE gates, etc) as well which will differ from model to model. It's not as
80.
by LarsDu88 3mo ago
Someone do this right now. I will join!
81.
by LarsDu88 3mo ago
A really good startup idea right now... Use kimi k3 to reproduce the kimi k3 asic design and start fabbing it immediately. In 12-18 months, start spinning up your own cloud and start competing with the frontier labs ASAP. Who needs superint
82.
by LarsDu88 3mo ago
The narrative that superintelligence is imminent is partially at fault here. There are competing definitions of what intelligence even is, and the one that I find most striking is from Francois Chollet which is that intelligence can be boil
83.
by LarsDu88 3mo ago
You're confusing Tesla for like 10 other Chinese brands. Tesla too is falling behind, particularly when it comes to price. BYD may very well be pulling ahead on battery tech and vertical integration right now. In a few years Tesla&#x27
84.
by LarsDu88 3mo ago
To be fair Unreal has rock solid netcode and Doom The Dark Ages has non-existant netcode.
85.
by LarsDu88 3mo ago
Unity does not make that much money from assets. They make the majority of their revenue through licensing their engine to "whales"... the small percentage of games that make huge revenue. They also make money through ad services.
86.
by LarsDu88 3mo ago
MSFT could have opened up idTech completely since they make 0 dollars from licensing the engine anyways. Microsoft's game divisions make money through making games, so opening up the engine itself would've been conducive to their
87.
by LarsDu88 3mo ago
Microsoft, one the world's greatest monopolists, bequeaths a game engine monopoly unto Epic Games, in one the biggest corporate blunders of all time. If they were smarter about this, they would commoditize their compliment and open sou
88.
by LarsDu88 3mo ago
Reddit thread with ongoing information regarding the layoffs: https://www.reddit.com/r/Doom/comments/1up5pta/95_reportedly...
89.
Majority of id software to be laid off by Microsoft(bsky.app)
34 points by LarsDu88 3mo ago|10 comments
90.
by LarsDu88 3mo ago
Scott Miller (founder of Apogee/3dRealms) stated that id Software and most of programmers (the team behind idTech and Doom: The Dark Ages) will be let go. Basically one of the last cutting edge engines not built/maintained by Epic

More ›