thehamkercat

4 ms·

1.
by thehamkercat 16d ago
I hate to break it to you, but this is not the first time snapchat has done this Example: https://www.bbc.com/news/world-europe-68099669 and there were many more
2.
by thehamkercat 17d ago
I think people shouldn't just jump to distros which are getting heavily hyped in media/Youtube, cachyOS had similar wave, and now Omarchy does. (example: NetworkChuck, Primeagen? and a few others) also, archlinux is much easier to
3.
by thehamkercat 19d ago
A vibe-coded password-manager? Sure! where do i sign up?
4.
by thehamkercat 19d ago
it _is_ vibe-coded, you can read the commit descriptions in sesame-desktop repo I wouldn't trust any password manager or critical applications like this written after 2024 bitwarden/KeePassXC are already more than enough
5.
by thehamkercat 21d ago
twitterwebviewer.com was registered in Hong Kong according to https://www.whois.com/whois/twitterwebviewer.com Looks fine, they don't care about the notices i guess
6.
by thehamkercat 1mo ago
I think you're confusing them with samsung who was caught in 2023 for this https://www.theverge.com/2023/3/13/23637401/samsung-fake-moo...
7.
by thehamkercat 1mo ago
They're probably scrambling right now to figure out how to block this
8.
by thehamkercat 1mo ago
Can you explain how does it work? like how is the previous K/V cache used when you switch to another model? Source?
9.
by thehamkercat 1mo ago
Muse Glimmer 30B seems to be on par with Qwen 3.6 27B (4 months old) but Qwen 3.8 27B is dropping this week...
10.
by thehamkercat 1mo ago
I personally think it's snake-oil marketing with all these smart-model-routing products/projects prompt-cache won't work with these
11.
by thehamkercat 1mo ago
> NeMo Switchyard, an open source library for smart routing > When deployed, NeMo Switchyard can intelligently direct each request to the most capable and suitable model for the job How do routers like this handle prompt caching when
12.
by thehamkercat 1mo ago
> 11.08× faster and generated tokens 16.36× faster than the same workload in the same stock VM. So this was the comparison, for me the title was a bit confusing
13.
by thehamkercat 1mo ago
From README: > On the 128 GB M5 Max, clean end-to-end image+audio and embedded-video+audio renders completed in 74.58 and 76.99 seconds respectively, each with about a 40.1 GB peak physical footprint and zero swaps. Looks like it uses 40
14.
by thehamkercat 1mo ago
This sounds cool
15.
by thehamkercat 1mo ago
"My phone is a server now" would've been better -- > built in battery backup Isn't it always recommended to remove battery if you plan to use a phone or a laptop as a server? ( Fire hazard ) or is it fine to just limi
16.
by thehamkercat 1mo ago
I believe it's because they are below $20 Million revenue limit (which Kimi K3's license has) So we won't see any price decrease unless Kimi changes the license of K3
17.
by thehamkercat 1mo ago
And now nobody seems interested in it because the price hasn't gone down it's still $3/$15 for all providers on openrouter because of some Kimi license https://openrouter.ai/moonshotai/kimi-k3#providers
18.
by thehamkercat 1mo ago
https://news.ycombinator.com/item?id=47969237 Posted 3 months ago as well, might not be new
19.
by thehamkercat 1mo ago
> 10 Million Tons of mass/year launched into orbit Not something to be proud of...
20.
by thehamkercat 1mo ago
> I almost do not need a traditional OS anymore. You sure about that?
21.
by thehamkercat 1mo ago
Why are companies slapping "OS" in their product naming? it's stupid
22.
by thehamkercat 1mo ago
I've only read a little about the Dunning-Kruger effect, but I think I understand it pretty well
23.
by thehamkercat 2mo ago
i think it should be "by adding stickiness to the initially chosen provider of that model"
24.
by thehamkercat 2mo ago
openai: "Our models are more dangerous" anthropic: "No, our models are more dangerous"
25.
by thehamkercat 2mo ago
nerfing*
26.
by thehamkercat 2mo ago
They also got a lot of goodwill by letting people use codex subscription in any harness they want, anywhere they want
27.
by thehamkercat 2mo ago
wait what? shouldn't it make you more hungry/angry? when the food doesn't come...
28.
by thehamkercat 2mo ago
Not a scandal ofc, but people use local models mostly for privacy and using a closed-source, VC-backed app that might change anything in the next update might not be best for privacy
29.
by thehamkercat 2mo ago
A friendly reminder that both LM Studio app and now this new LM Studio Bionic app are closed source. Since most people are unaware of this fact.
30.
by thehamkercat 2mo ago
Hey, you can disinfect with it 3 times to get 120%

More ›