Viewing profile — chorizo

Light dependent resistor, opamp and relay. No digital components required.

They tweeted that ARC-AGI-3 results take longer to run, so we’ll need wait a bit longer.

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.

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 f…

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 ima…

The big benefit is ROM cells require fewer components than DRAM. So the chips would be tiny, dense, cheap and consume far less power.

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 o…

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-s…

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/

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 4x…

I’m developing open source tools, so happy to have all my context traces be public if it helps improve future models.

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.

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 th…

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.

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 …

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.

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.

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 commercia…

> 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 …

Not to mention Nvidia loans out their engineers to academic computational software teams to help build CUDA ports.

What we already have is uncontrolled geoengineering. Having said that, geoengineering might lead to geopolitical conflict. The current situation is making the tropical/equatorial a…

And the advanced planets looked like the SFU campus.

As someone who lives in Vancouver, that’s a huge relief. I won’t have to adapt to new biomes once we go offworld.

That’s 24GB VRAM. Not enough to run a 27B model at a useful quant+context size.

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…