SwellJoe

11 ms·

241.
by SwellJoe 3mo ago
I do not. He spends his time working on a language used by and supported by several startups and a few major corporations, a language that seems heir apparent to C, at least among a lot of developers. The ICO in your name leads me to believ
242.
by SwellJoe 3mo ago
WTF are you talking about? He made Zig and is employed by the Zig Foundation, of which he is founder and president. He can't lose his job because he ruffles feathers.
243.
by SwellJoe 3mo ago
He shouldn't talk about his own experience? Why not? Why is it in poor taste to talk about a difficulty he went through?
244.
by SwellJoe 3mo ago
So, announcing the fallback is better than doing it silently, but the fact that Fable falls back frequently for the kind of work I do (a lot of security oriented stuff lately, but it falls back on seemingly random stuff, sometimes, too),
245.
by SwellJoe 3mo ago
I have no idea. Seems like normal stuff. I used Kimi Code with K3 to add support for Kimi Code to flar ( https://swelljoe.com/post/i-let-every-agent-implement-its-ow... ), a task I've done with almost every major mo
246.
by SwellJoe 3mo ago
I got the $19 plan, and it's anemic. One tiny task blew through the 5-hour budget and 19% of the weekly budget. A completely useless amount of usage. OpenAI's $20 plan feels like 100x more generous (I don't think I'm exa
247.
by SwellJoe 3mo ago
I don't understand how the K3 numbers keep coming out cheap for people. I recently started to add it to my security auditing benchmarks and found it was going to cost about twice as much as Opus 4.8. It blew through the $100 budget I&#
248.
by SwellJoe 3mo ago
I certainly don't see much in the way of cached token rates when I use the API, whereas when I use DeepSeek via API, it's mostly at the cached token rate.
249.
by SwellJoe 3mo ago
Two things can be true: 1. LLMs are remarkable and they've uncapped a supply/demand loop for software that has previously been much more tightly constrained than anyone realized. It turns out that if software is much cheaper and f
250.
by SwellJoe 3mo ago
There's no need for insults.
251.
by SwellJoe 3mo ago
Definitely, but there's also a distinction between "weights available" and "you can do stuff with those weights". There are models you can download but the license prohibits redistribution, fine-tuning, etc. The ear
252.
by SwellJoe 3mo ago
Sorry, I took "No, but" as a disagreement. You're right I was unnecessarily snarky about it.
253.
by SwellJoe 3mo ago
A subagent running doesn't mean that much, though. I often walk away while the subagents are chewing.
254.
by SwellJoe 3mo ago
"No, but a small university team with dark time on the state school system's cluster can build a basic, functional LLM." Hey, remember when we were just talking about how it "needs public funding and reputable institutio
255.
by SwellJoe 3mo ago
Is that actually happening, though?
256.
by SwellJoe 3mo ago
I'm not at all disagreeing with you about calling these models "open source". They mostly aren't and we should stop conflating the two things.
257.
by SwellJoe 3mo ago
I don't know what you mean, so, if I am obtuse it is from a genuine lack of understanding. What is obtuse about that statement?
258.
by SwellJoe 3mo ago
"Compiling/creating the Linux kernel or Chromium is no child's dance either" I can compile the Linux kernel on a modest 15 year old laptop. Sure, it took 35 years and thousands of people to build it into what it is today
259.
by SwellJoe 3mo ago
Keeping a conversation with a very large model active requires hundreds of GB of memory. If my conversation can never be swapped out, like when I go to lunch or take an afternoon walk, that's several hundred GB not available for other
260.
by SwellJoe 3mo ago
While Anthropic is far behind on caching and efficiency (made more dramatically apparent by how much cheaper it is to use GPT 5.6 Sol at API rates than even Opus 4.8, much less Fable), a bunch of people forcing their way to the front of the
261.
by SwellJoe 3mo ago
This is just making it more expensive for everyone else, right? How Claude handles its sessions is none of my business. I'm going to let them do the best they can to provide good service for everyone, and if they can't/won&#x
262.
by SwellJoe 3mo ago
Open Source AI is challenging. Even tiny models require tremendous resources to produce, so all the things we know about "open source" for software, like a random person in Nebraska can produce a critical piece of the world's
263.
by SwellJoe 3mo ago
The way Anthropic is playing the "AI is scary and dangerous and only we can be trusted with it" game is clearly aimed at regulatory capture, which would only be good for Anthropic. It's worrying how cozy some of the tech lead
264.
by SwellJoe 3mo ago
I've tried it, but, it's not reasonable speed, at all. It's 9-13 tokens per second, which is not usable interactively and not worth using for long-running API stuff when DeepSeek V4 Pro is so cheap via their API. Laguna S 2.1
265.
by SwellJoe 3mo ago
The Amiga was the best and worst thing to happen to a kid in the late 80s. It was heartbreaking watching Commodore die, but it was an amazing introduction to the possibilities of computing. I made a ton of music with OctaMED, learned a bunc
266.
by SwellJoe 3mo ago
That explains the very old models mentioned on the page. There are models you can run locally on anything that would be more useful than llama 2 70B. e.g. the new Prism Bonsai 27B ternary model is 6GB, runs on any modern Mac, any PC with
267.
Antares from Cisco: Highly Efficient Open Models for Vulnerability Localization(blogs.cisco.com)
2 points by SwellJoe 3mo ago|0 comments
268.
by SwellJoe 3mo ago
I think they're distilling "reasoning", not merely code. There's plenty of human generated code. What they're trying to extract is the process by which really large models "think" their way through complic
269.
by SwellJoe 3mo ago
"It is already understood that you own the output." I don't think it's settled that anybody owns the output. There seems to be some question whether LLM output can be copyrighted (and there should be). I'd rather
270.
by SwellJoe 3mo ago
Every bike is a balance bike. That's how bikes work.

More ›