julianlam

5 ms·

1.
by julianlam 4d ago
Six months ago this ceased to be the case for some (if not most), and the quality of LLM generated code has increased over time. I'm sorry to be the bearer of bad news: human coding has not improved a lick since then.
2.
by julianlam 5d ago
At a high level, we're still at the stage of AI development where we're taking cues from nature. Take the most recent qwen and deepseek models with offloadable n-grams, which function (both in name and vaguely in capability) like
3.
by julianlam 7d ago
Can confirm. I've a single R9700 and have maxed out at 45 tokens/second on llama.cpp with Q4 Qwen 3.6 27B (with MTP)
4.
by julianlam 7d ago
The way this comment was going I was pleasantly surprised to see that the iPhone Duo adhered to the ISO standard.
5.
by julianlam 16d ago
As a thin client to access cloud models this is an astounding waste of money.
6.
by julianlam 16d ago
> local models are close to the point where, with a sufficiently well-architected harness, you can get results that are on par with the experience you'd have with cloud inference In my opinion, 98% of the work most devs would send t
7.
by julianlam 16d ago
16GB VRAM could load a small quantised qwen 27B model but it would be a ways away from a frontier cloud model. Though keep in mind not being beholden to shenanigans from said cloud companies (and interference from government entities!) is d
8.
by julianlam 17d ago
> That’s less than 164USD worth of GLM5.3 tokens on the inference market. I can cherry pick stats too. The other day I heard mention of someone paying $200/mo for Claude Code. At those rates my local LM setup pays for itself in a si
9.
by julianlam 19d ago
I run my agent (pi) in a podman container. I have poked a couple directories in (.git, .pi, couple others) to ease some pain points but otherwise my home directory is inaccessible. It seems to be a low-effort compromise to running a VM or s
10.
by julianlam 21d ago
When you hear stories like "Opus 5 thought for 20 minutes and then denied my request" it really puts wind in this sails of Local LMs
11.
by julianlam 27d ago
I experienced something similar with the AliBaba app. Whenever I had it installed, my Bluetooth devices would only ever send audio in "headset" mode — very poor quality. Uninstalling the app fixed it up.
12.
by julianlam 27d ago
You're likely using 3.6-35B-A3B, the 3.8 is currently a 27 billion parameter dense model.
13.
by julianlam 1mo ago
Been that way all along.
14.
by julianlam 1mo ago
Sounds like the Republican party would really support the adoption Newspeak. > It was intended that when Newspeak had been adopted once and for all and Oldspeak forgotten, a heretical thought — that is, a thought diverging from the princ
15.
by julianlam 1mo ago
Define "easily". My laptop (2023) was configured with 16gb. Getting to the point where I was able to run a 30B model required $500 in memory.
16.
by julianlam 1mo ago
Wonder if it's possible to share a common cache via cachyllama between 27B and 35B-A3B
17.
by julianlam 2mo ago
> Ofc like you mentioned it'll sometimes lead to misleading content (eg made up food pictures in menus), This does happen without AI already. My local shawarma joint has a deal of the day poster of a chicken shawarma that's obv
18.
by julianlam 2mo ago
Spot on. The same arguments pop up when new therapies are unveiled for various sightedness disorders, hearing impairment, etc. How _dare_ you impugn upon my identity that has been reinforced through my upbringing constantly? Etc.
19.
by julianlam 2mo ago
Yeah, I do the same. Soup sock in a vacuum cooker. It's the chinese way to make soup. The chicken does turn out dry. Perhaps it's worth taking the chicken mear out (leaving the bones) after an hour and adding it back after the sou
20.
by julianlam 2mo ago
What an Americentric point of view. Not everybody measures weight in pounds, and to the primary readership of the BBC, stones and kilograms mean much more than weighing pounds. Source: Canadian who understands neither kg or stones.
21.
by julianlam 2mo ago
Why do prescription medications cost so much, and generics so little (comparatively)? Artificial inflation to recoup R&D.
22.
by julianlam 2mo ago
The ones in Canada usually have sloped roofs to block rain. It's enough. Indoor ones obviously don't need it.
23.
by julianlam 2mo ago
Why Tangled instead of something more established like Codeberg, or if f/loss, Forgejo or Gitea? Just because ATProto vibes?
24.
by julianlam 2mo ago
Forever soup isn't new, of course. Poor families do it to cheaply make long lasting meals. My late maternal grandmother used to have a pot of forever soup on the stove, and she would put whatever she had on hand in. Spare ribs, check.
25.
by julianlam 2mo ago
The floating ring of space debris around Earth would like to have a word with you.
26.
by julianlam 2mo ago
Every time this sort of news pops up, I think "why aren't they incentivising and subsidizing fertility treatment"? You want to raise the fertility rate, yes? You could convince an entire population to have more babies the har
27.
by julianlam 2mo ago
Yeah Hopper didn't start off as a booking site. Just a private group not affiliated with any airline group. :)
28.
by julianlam 2mo ago
Sounds like Hopper, but AI. Why not just use Hopper?
29.
by julianlam 3mo ago
You wouldn't distill a car.
30.
by julianlam 3mo ago
[citation needed]

More ›