drob518

5 ms·

31.
by drob518 2mo ago
Yea, I felt like I was perpetually at the top of a roller coaster and going over the hump, but never finding the bottom. And I like roller coasters. But no thanks. Not even sure what the benefits of this view really are.
32.
by drob518 2mo ago
Well, there goes the neighborhood. This is why we can’t have nice things.
33.
by drob518 2mo ago
I’m sure someone does, but I’ve never seen anything other than vague statements like Anthropic’s “inference is profitable” comment. I suspect everyone is playing everything close to the vest because they aren’t yet public and they want to c
34.
by drob518 2mo ago
I haven't noticed the Chinese models going up in price for the same model. They do release new versions of the models with different prices that are higher. But everybody is doing that. One fine point is that deepseek-v4-flash-0731 is
35.
by drob518 2mo ago
Of course. Pricing is always changing, but typically it goes down over time, not up. So, if you're showing artificially low pricing from the start based on a teaser rate, IMO, you shouldn't be using that to show where you appear o
36.
by drob518 2mo ago
Hm. GLM is more expensive in all dimensions than DS but it has a lower weighted average input? How is that?? Something seems off. EDIT: Looks like they are swizzling around the pricing dynamically on that page, on both the GLM and the DS si
37.
by drob518 2mo ago
We need to figure out what the real pricing is for a going concern. Right now, everyone is subsidizing and discounting to grow (or maintain) market share. The big question is whether the steady state, market derived inference pricing is abo
38.
by drob518 2mo ago
Seems disingenuous to draw frontier graphs with starter pricing.
39.
by drob518 2mo ago
Yep, for cpu I’m sure that’s true. It isn’t often that I wish the M1 was faster. But I need more memory. 16 GB was fine five years ago, but it’s getting cramped for me now.
40.
by drob518 2mo ago
Yea, that wasn’t going to happen.
41.
by drob518 2mo ago
I’m milking my M1 Air, too. It’s still going strong, and I just can’t justify replacing it. Would love 64GB or more memory, though.
42.
by drob518 2mo ago
I use Pi and I’m coming to the conclusion that mostly all I need is a bash tool.
43.
by drob518 2mo ago
Two words for you: Kobayashi Maru
44.
by drob518 2mo ago
Well, it depends on how they deliver that savings. If it’s through something unique and defensible, something that only they (with some moat) can do, then yes, they can argue for a fraction of the savings. But if prices just go lower becaus
45.
by drob518 2mo ago
So… they’re both middlemen. And that’s enough to justify a $7B acquisition???
46.
by drob518 2mo ago
> A company like OpenRouter that promises the cheapest tokens by the minute becomes essential on the low-cost margin. Okay, I can see that, but if their value is just cost optimization, their ability to mark up the tokens becomes increas
47.
by drob518 2mo ago
I don’t understand it, but good for OpenRouter. Personally, I think OpenRouter’s value is somewhat perishable. In the early AI market where there are a jillion models and anew one every week, it’s great to be able to try them all without ha
48.
by drob518 2mo ago
Hm. This seems random. Other than the fact that they are both technology companies, is there anything else that they have in common?
49.
by drob518 2mo ago
Wow, move the goalposts much? Okay, I’ll stipulate that somebody, somewhere, at some time might have been promoted to get them out of the way. But I’ll tell you it’s definitely not the norm in any company that I’ve ever worked in, to the po
50.
by drob518 2mo ago
Pi -> OpenRouter -> Deepseek V4 Flash running on a random 3rd party provider (not Deepseek itself)
51.
by drob518 2mo ago
Okay, but the hiring manager didn’t hire him because he was a poor performers, right?
52.
by drob518 2mo ago
I think they do the random thing.
53.
by drob518 2mo ago
Yep, I’ve seen that, too. I’ve also seen people give up a promotion after they decided that managing a team wasn’t what they were good at. A wise person understands themself and knows not to chase the brass ring if that’s not actually what
54.
by drob518 2mo ago
Yep, agreed. I have seen that, where engineers decide they really want to be something else and it actually works for them.
55.
by drob518 2mo ago
I’ve been a VP at F500 companies and never seen it done deliberately. There are plenty of ways to manage poor performers out of an organizations. Promotion isn’t one of them.
56.
by drob518 2mo ago
I have had it loop once. I had to abort the turn, clear the context (maybe I just compacted it?) and restarted. That seemed to clear it up. In general, I have not found that V4 Flash 0731 loops a lot or even any more than other models.
57.
by drob518 2mo ago
I’ve never found this to actually be the case (that somebody promoted someone they thought was incompetent to “limit damage”). Yes, incompetent employees get promoted sometimes, but it’s typically because they have been brown-nosing an exec
58.
by drob518 2mo ago
For cost reasons, I’ve been using Flash for the reviewer, too, but I plan on trying to use Pro for that. Thus far, however, Flash has been doing well at reviewing. I’m cheap as I’m paying for all the tokens myself.
59.
by drob518 2mo ago
IMO, ELO rating from The Intelligence company and arena.ai are more representative of rankings since they use humans to judge a head-to-head comparison between a couple models at a time. https://www.intelligence.ai/ http:&
60.
by drob518 2mo ago
Yep, exactly. I keep saying that I want the “coding expert” extracted from these multi-T parameter models to run locally on reasonable hardware (large laptops, not servers). Yea, I know there’s no single “coding expert” that you can actuall

More ›