Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
chorizo
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
31.
▲
by
chorizo
2mo ago
They tweeted that ARC-AGI-3 results take longer to run, so we’ll need wait a bit longer.
32.
▲
by
chorizo
2mo ago
This also reflects my experience and should put to bed the distillation rumours. This model feels nothing like the Claude models, including tone and blindspots.
33.
▲
by
chorizo
2mo ago
And the customers can wait for the new phone released next year. These are edge models - the average customer doesn’t need the latest frontier model. Just needs to be good enough for the features you promised.
34.
▲
by
chorizo
2mo ago
Yes, each rom bit can be a transistor or even a diode with a decoder circuit. Simplest Dram cell is capacitor+transistor - and you need a clock, refresh circuit etc. Someday, I imagine model weights could even be encoded as analog resistor
35.
▲
by
chorizo
2mo ago
The big benefit is ROM cells require fewer components than DRAM. So the chips would be tiny, dense, cheap and consume far less power.
36.
▲
by
chorizo
2mo ago
That’s not going to be true forever. As models mature, we will hit diminishing returns. Major improvements will come annually rather monthly - matching the roughly annual release of new processors. Model ROM’s will likely get integrated int
37.
▲
by
chorizo
2mo ago
Baking the base models on to ROM makes a lot of economic sense. SRAM for the KV cache & fine-tunes, not so much. Sure you’d get incredible speeds but it’s not scalable from a die-size or cost perspective. Rather base model on ROM + KV c
38.
▲
by
chorizo
2mo ago
On a browser, you can strip the share id and post title to create a “clean” old.reddit link, with this pattern: https://old.reddit.com/r/bapcsalescanada/comments/1vd9h0j/
39.
▲
by
chorizo
3mo ago
I’ve heard that too, but since it’s a stealth model, I suspect they change it to whatever preview model provider that offers a free endpoint. In the past few months, I’ve gotten 4xx errors identifying the provider as DeepSeek.
40.
▲
by
chorizo
3mo ago
I’m developing open source tools, so happy to have all my context traces be public if it helps improve future models.
41.
▲
by
chorizo
3mo ago
Starting to wonder if the free big pickle model on opencode has been DSV4F0731 for the past few months. It’s been incredibly fast and good.
42.
▲
by
chorizo
3mo ago
I still use a 16GB MacBook Air for all my development and simulation work (4P+4E+8GPU cores). Even wrote my own Metal gpu code to offload part of the simulation workflow. Modded the MacBook Air with thermal pads and a homemade cooling pad m
43.
▲
by
chorizo
3mo ago
Certainly not demanding they open source any IP that might give their competitors an edge. Rather I’m saying that pulling a gpt-oss-120b would a good PR move right about now.
44.
▲
by
chorizo
3mo ago
Or even more poignant, have terra write code and it will find lots of bugs in its own code. But let it iterate code reviews/fixes/test cases a few times, and you’ll have something nice. The only cases where weaker models fail en
45.
▲
by
chorizo
3mo ago
At this point, Anthropic releasing a sonnet 4.x class open weights model would be a positive PR move that doesn’t affect their bottom line.
46.
▲
by
chorizo
3mo ago
That’s a really good point. Folks really need to read the papers coming out of these Chinese labs. Every paper from the DeepSeek team has been a step change.
47.
▲
by
chorizo
3mo ago
Appreciate this point. Back in grad school, I published a peer reviewed paper with all the source code and datasets used. Got heckled at a conference talk by staff from a commercial lab. They shouted, we figured this out five years ago, lol
48.
▲
by
chorizo
3mo ago
> Everything you know about 6800 assembly will not make your note-taking app any better. For most apps, it makes sense to profile your app for expected use cases. Put even minimal thought into making the hot codepaths faster by moving un
49.
▲
by
chorizo
3mo ago
Not to mention Nvidia loans out their engineers to academic computational software teams to help build CUDA ports.
50.
▲
by
chorizo
3mo ago
What we already have is uncontrolled geoengineering. Having said that, geoengineering might lead to geopolitical conflict. The current situation is making the tropical/equatorial area less habitable while creating overall benefits for
51.
▲
by
chorizo
3mo ago
And the advanced planets looked like the SFU campus.
52.
▲
by
chorizo
3mo ago
As someone who lives in Vancouver, that’s a huge relief. I won’t have to adapt to new biomes once we go offworld.
53.
▲
by
chorizo
4mo ago
That’s 24GB VRAM. Not enough to run a 27B model at a useful quant+context size.
54.
▲
by
chorizo
4mo ago
I built a little research dashboard that monitors new papers from specific research labs. There is a paper ingest skill that writes Python scripts using pdfplumber to dismantle pdf’s. I have also used it to fetch supplementary information t
55.
▲
by
chorizo
4mo ago
I extensively use open source Chinese LLM’s for coding. Reading the reasoning traces, especially when planning and debugging code, is valuable. I will pause the llm when immediately when it’s made incorrect assumptions. Often I see I mentio
56.
▲
by
chorizo
4mo ago
Safari reader mode on mobile works great. But then again, this is a site where you should not need it.
57.
▲
by
chorizo
4mo ago
Organized/scripted scalping isn’t new at all in the Western world. Just look at concert/game tickets. It assumed that if you really need it, you’ll alarm clock the release and even then might end up paying a scalper.
58.
▲
by
chorizo
4mo ago
In Valve’s, they’re raffling the opportunity to make a purchase. For a generalized Vickery auction, it’s assumed the buyer will complete the bid, but that’s not so in valve’s case. What’s a good solution when there is a significant % of bid
59.
▲
by
chorizo
4mo ago
I made the same switch. Impressed with the sheer value of Deepseek v4 Pro and Minimax M3 so far. I mostly work on an open source academic simulation tool, so I’m happy to be a source of training data.
60.
▲
by
chorizo
4mo ago
Imagine security cameras everywhere. Oh wait, we already have that. Imagine llm’s analyzing every frame of every security camera everywhere (including shared ones like stores, flock, ring etc). Then an agentic judge issues warrants.
More ›