

I know what I said.


I know what I said.


Because their extremely dangerous and 100% independent, unprompted, rouge Sam Hypeman model needs more RAM. Duuh…


This has no relation to technology whatsoever. Neither Meta or zuck are relevant to anything besides spreading misinformation and interfering with elections around the world. As for the other clown, he’s famously quoted by “it’s all computers!”, and single handedly decided to change name of AI for some recently developed unknown namechanging kink.
Ah, I was worried that there was some local model distro meta that I wasn’t aware of. I have no knowledge of AMD but according to DeepSeek we seem to have somewhat comparable hardware for local LLM. I just noticed the 64GB RAM and figured you’d be on a whole different level, but apparently that setup is in the mid-high tier, spanning from 7-32B? What kind of models have you tried so far, and did any stand out? I’ve only been able to get a 13B Qwen3.5 model as most myself (which was way too slow to be worth it). From what I have picked up >24B is where you get a generally good model that goes beyond a niche purpose.


What are you even on about… Didn’t you receive your stim check, personally signed from big daddy himself already? Ya’ll would’ve received another $5000 bribe payout! Now try asking dems for the same!
How is switching OS related to local models? Are there any OS that is more efficient? Is there some software that serves certain OS better? I’m curious because I’m also using Mint, have been for about a year now after ditching Windows and I was actually exploring potential benefits of other distros for the first time earlier this week. I’d love to learn if there is something to keep in mind, or is it related to your AMD hardware?


Can’t believe they forgot to add “make no mistakes”


This clearly staged photo op is a interesting choice. I can see how it’s meant to humanize Sam, and portray him as a common, ordinary person. But why does it looks like someone just gave him a compliment and made him flustered?


I doomscroll Lemmy /and/or my most recent find which I found to be an excellent source for AI/LMM related news that’s actually interesting, hackernews. I also keep a radio plugged in next to my bed, with one of the public service radio channel tuned in 24/7. During the night they usually do reruns of various educational shows that aired during the day, no music, no advertisements, just dull, calm voices speaking about some mildly interesting topics, like gardening tips for example. I lower the volume a bit so I have to really focus if I want to actually listen, otherwise it’s just comfortable noise. It helps a bit with my usual overthinking that usually hits whenever I go to bed.
I’ve also dabbed my toes in some ASMR and podcasts, and while I found a couple that sort of worked. The tradeoff is that I have to use my phone speakers to listen to anything streamable, which includes me rolling and tossing during the night, and my phone either ends up under me, or has been tossed around so much that I can’t find it by the next morning. Often with a drained battery too, which is just an annoying way to start your day.


Thanks for the share! Appreciate actual, non-toxic discussion about LLM usage on Lemmy.
The reason is money, not creativity. If the money spent on tokens doesn’t yield a profit, nobody would be releasing 5 games a week.Edit: didn’t realize that it had already been mentioned. I’ll change to something else to avoid parroting;
You should give it a try and see find out yourself. The issue is not that people vibecode their own games, who cares. Let them enjoy themselves. I believe that the fun part is watching your game come alive and tweaking it to your preference. The issue is rather that a lot of people choose to publish them, which I suspect is because of the potential profit it may have. My question is why aren’t there a website where people can share, or even sell their vibecoded games for those interested? That would benefit all since sharing assets, prompts, MCP, and what not, would lower cost of spent tokens.
Edit 2: A quick search answered my question. Apparently there are multiple websites dedicated to serve this purpose already.