ComputerGuru

6 ms·

31.
by ComputerGuru 29d ago
But the discount is only available via OpenRouter.
32.
by ComputerGuru 1mo ago
Does OpenRouter eat this cost to get their hands on a copy of the conversations people are using with the model?
33.
by ComputerGuru 1mo ago
You can’t check it with only the algorithm, you need the secret seed key. Which will be different for each provider (and they’ll probably have and use multiple). And you need the llm itself, to generate the potential tokens at each step.
34.
by ComputerGuru 1mo ago
This was 3.5 flash lite, actually, and after prompt tuning. It was very clearly an issue that correlated with input (JSON array) size, the more elements in the batch, the higher the error rate. 3.0 flash (not lite) handled it like a champ t
35.
by ComputerGuru 1mo ago
We have 3.7 Flash now, actually, and it costs just a hair over the old 3 Flash Preview while being better!
36.
by ComputerGuru 1mo ago
Speaking from experience here, flash lite models have amazing price, speed, and perform far above their size, but are susceptible to very bad instruction following and recall when either complexity or context size inch up. They’ll just forg
37.
by ComputerGuru 1mo ago
Complaining about overthinking in xhigh then pointing out output had bugs with thinking turned off seems like it’s missing the obvious compromise?
38.
by ComputerGuru 1mo ago
Thanks for the corrections, but I am specifically talking about semantics from the Unicode Technical Committee's perspective, of the underlying Unicode codepoint(s). There is a reason some end-user-viewable glyphs can be formed in mult
39.
by ComputerGuru 1mo ago
It’s really a complete and total nothing burger. Extra code points were added, might have been an issue when we were trying to cap the total number below needing some arbitrarily fixed number of bytes for convenience, but now that’s no long
40.
by ComputerGuru 1mo ago
iOS at least renders it as an n with dieresis; is that how it was intended (I’m unfamiliar with musical notation)? If so, what is so difficult about it? In fact, (semantics aside, from a technical perspective) the preference should always b
41.
by ComputerGuru 1mo ago
Clickbait post, already made the rounds on in other sites and was mercilessly torn to shreds. The answer is that millions do in production. PgBouncer is only needed for stateless backends, and even then, only under specific circumstances.
42.
by ComputerGuru 1mo ago
I never trust OpenRouter to forward parameters correctly and would only ever conduct benchmarks with the official api, personally.
43.
by ComputerGuru 1mo ago
At this size, it certainly doesn’t have to be.
44.
by ComputerGuru 1mo ago
But this isn’t a query that should need to be forwarded to the cloud for acting on!
45.
by ComputerGuru 1mo ago
Can you share more about the architectural/design tradeoffs you considered or decided upon? Particularly for me, why is a model that is intended mainly to just make tool calls and marshal the results back focusing on speed? Speed as an
46.
by ComputerGuru 1mo ago
We are still paying the price for that today, some of us a lot more than others. It was well-intentioned but with hindsight being what it is, historians pretty much agree that it was the cause of a lot of the problems we fight with today.
47.
by ComputerGuru 1mo ago
Yes, I found that as a workaround but I can't help but feel it's maladaptive since your spine is no longer aligned.
48.
by ComputerGuru 1mo ago
What an awful website for such a wonderful feature. I don’t expect to see this in the original Find My (Apple’s) as their implementation is fully custom and works altogether differently (and already had a protocol-level rewrite not too long
49.
by ComputerGuru 1mo ago
I (lifelong side sleeper) started using a (dedicated) knee pillow about a decade ago and I am completely torn about them. On the one hand: amazing. Better spinal alignment, zero sensitivity or soreness where the knees meet, better sleep, e
50.
by ComputerGuru 1mo ago
There is no good reason to believe language-specific models are going to be any meaningfully smaller, just worse. Same as English-only models vs those trained on a multilingual corpus.
51.
by ComputerGuru 1mo ago
Just to play devil’s advocate: you can’t compare Qwen to a (proprietary/closed source) hosted model and deduce that Qwen is overthinking, as Qwen gives you the full reasoning/thinking trace while all the proprietary models now giv
52.
by ComputerGuru 1mo ago
The tokenizers are included in the open s̶o̶u̶r̶c̶e̶ weights releases; you wouldn’t be able to use the weights without the corresponding encoder/decoder, in fact.
53.
by ComputerGuru 1mo ago
Look into tursodb for this.
54.
by ComputerGuru 1mo ago
What quantization level is that? Because official endpoints are slow .
55.
by ComputerGuru 1mo ago
I would have the same concerns but note that backup generators aren’t just run when needed but also have a weekly/monthly quota of hours they need to run for both testing and keeping them in shape.
56.
by ComputerGuru 2mo ago
What is MMD? That is not an svg.
57.
by ComputerGuru 2mo ago
Does the page actually load for anyone? I get stupid spa skeleton spinners.
58.
by ComputerGuru 2mo ago
It gave the svg for the wheels in the reasoning trace then forgot to include them in its final answer. Lol.
59.
by ComputerGuru 2mo ago
They could certainly still achieve that even if they had a slightly more expensive tier with a (pinky promise) "no training" bit set, or even if it were just an opt-in most users wouldn't bother with.
60.
by ComputerGuru 2mo ago
If only they would let you opt out of training use, it might actually be a viable option.

More ›