swalsh

5 ms·

1.
by swalsh 5d ago
So what's the felony benchmark at now?
2.
Show HN: Possess, a TUI to browse and transfer coding sessions(github.com)
3 points by swalsh 9d ago|0 comments
3.
by swalsh 13d ago
I was thinking about canceling my claude max sub after a few bad experiences. Kept hitting my usage limit, the quality of code seemed worse than Sol. This just made my decision. I'm moving to Codex Pro.
4.
by swalsh 14d ago
Meetings, research, look over work done etc.
5.
by swalsh 15d ago
It's not insulting. Not every problem is a frontier problem.
6.
by swalsh 15d ago
Now that it's a solved benchmark, can we get the animated version?
7.
by swalsh 15d ago
I've recently been running these agent sessions on more and more long running tasks because these latest models can do a REALLY good job on big chunks of work, and i've been watching them way less. It's starting to occur to
8.
by swalsh 1mo ago
For as long as you're on it, but eventually you run out of body fat.
9.
by swalsh 1mo ago
Calling it now, the downside is the effects are most apparent on higher dosages, but you cant take higher dosages forever. Eventually you run out of body fat, have go reduce the dose, and the benefits dither.
10.
by swalsh 1mo ago
You can probably get a prescription for a low dose, but you'll be self pay.
11.
by swalsh 1mo ago
Yes they run in parallel via LMStudio (250k context)
12.
by swalsh 1mo ago
yeah, i consider that slow.
13.
by swalsh 1mo ago
WOW, my first try running on my 2 3090's, it was a bit slow... but it FEELS like opus 4.5, i gave it an image and a broad overview of what I wanted it to build, and it built the whole thing from beginning to end.
14.
by swalsh 1mo ago
I suspect mythos demonstrated a fully autonomous offensive hack in a similar way Open AI's models performed, and the government is reacting to it the same way we reacted to blackhat. The threat is real.
15.
by swalsh 1mo ago
Yeah, so I'm going to confess. Ive built production AI systems that have very specific jobs, deployed them, and moved on. The cost is not noticeable, i had the system dialed in. Not worth the effort to reevaluate a newer model to se
16.
by swalsh 1mo ago
Lol perfect analogy. I'm still paying for claude because the quality is unmatched.
17.
by swalsh 1mo ago
I think profitability is a matter of accounting. Inference is where money is made, but training is where money is spent. We keep getting new models every few months, but frankly the old models are still quite usable. I suspect labs will s
18.
by swalsh 1mo ago
Its tough to go from max account at home and pay per usage enterprise account at work with heavy usage limits... but the limits are there because pricing is insane. Feel like I'm in the $5 Uber rides phase at home.
19.
by swalsh 1mo ago
That seems to be the one unchanged variable of time.
20.
by swalsh 1mo ago
If this was a design firm, it might matter. But this is mostly a hiring ad for engineers, and a landing page for VC. I'd judge them more if they actually put effort into it.
21.
by swalsh 1mo ago
let me rephrase that: "The site itself demonstrates the team is spending their money in the places that matter, and using quick solutions for the stuff they need but isn't mission critical"
22.
by swalsh 1mo ago
By the middle of the 2030's the world we live in will be unrecognizable.
23.
by swalsh 2mo ago
The classifier for biology is so broad it makes me wonder what kind of stuff mythos was generating. Anthropic is known to be a bit dramatic, but they wouldn't have released something this broad unless they saw the model cross a signif
24.
by swalsh 3mo ago
Okay, work with me on this, because its a stretch. The new businesses created, might look very much like the old businesses that disappeared. But they're not focused on scale and profits. They're focused on an ideal of pushing t
25.
by swalsh 3mo ago
I have a composition scale, and regularly get vitamin d, and calcium tests (had issues with both) since taking the wegovy pill in Feb, I have not loss any bone density, and pretty minimal amount of lean muscle (i think any weight loss will
26.
by swalsh 3mo ago
This would be amazing for some of our "real-time" workflows, that need to fallback to AI for one reason or another. What used to happen is a rules based system did the majority of work, and occasional corner case would fall back
27.
by swalsh 3mo ago
This offends me, because I believe code is speech, and speech should not be restricted.
28.
by swalsh 3mo ago
Imagine the East India Trading company advertising the TAM of the new emerging markets in the early 1600's. I bet the numbers compared to the old world would be enormous, and equally unbelievable. IF, and yeah it's a gigantic rea
29.
by swalsh 3mo ago
There's issues where bad people with bad intentions use AI to do bad things. But to an extent the guardrails already built-in are sufficient. The real danger in AI is how it impacts society, our economy, and our perception of self wo
30.
by swalsh 3mo ago
The distilled versions miss the spark of the model. Its like they land in the uncanny valley of models.

More ›