Skip to main content
PPlamenu
HomeTrendingLive feedsPeopleGroupsRulesStaff
Sign in
PPlamenu
HomeTrendingLive feedsPeopleGroupsRulesStaff
Sign in

posted in Technology

eicker@eicker@lemmy.world
⁨20⁩h

Other AI Companies Should Be Very Afraid of What Just Happened to Character.AI, According to Its Own Bankruptcy Documents: Be afraid.

futurism.com/artificial-intelligence/ai-companies-afraid-characterai-bankrupcy
1810374
Open original page
BoostsQuotesFavs
TechAnon@TechAnon@lemmy.world
⁨4⁩h

Replying to @⁨eicker@lemmy.world⁩

I just got Qwen 3.8 Flash Next running locally. It’s damn good on an older machine with a newer GPU. Local LLMs should scare them the most.

⁨Oct⁩ ⁨15⁩, ⁨2026⁩, ⁨21:35⁩en
2000
Open original page
BoostsQuotesFavs
ssfckdt@ssfckdt@lemmy.blahaj.zone
⁨4⁩h

Replying to @⁨TechAnon@lemmy.world⁩

Yeah but who’s making the models?

3000
Open original page
BoostsQuotesFavs
crispbacon99@crispbacon99@lemmy.zip
edited⁨3⁩h

Replying to @⁨ssfckdt@lemmy.blahaj.zone⁩

They’re easily reproducible as copying and pasting. Then given you have a good enough graphics card you can just run it on your local pc completely offline.

0000
Open original page
BoostsQuotesFavs
TechAnon@TechAnon@lemmy.world
⁨3⁩h

Replying to @⁨ssfckdt@lemmy.blahaj.zone⁩

Many groups are making models now. I see a future where we have open source data sources, models, and weights. Niche groups can extend models to be trained on whatever they want.

0000
Open original page
BoostsQuotesFavs
DupaCycki@DupaCycki@lemmy.world
edited⁨3⁩h

Replying to @⁨ssfckdt@lemmy.blahaj.zone⁩

Mostly Chinese companies that aren’t necessarily required to generate infinitely compounding profits just to continue existing.

Edit: Also third parties such as volunteers who fine tune models for specific purposes. Just like with all software - everyone can learn and contribute when it’s open source.

0000
Open original page
BoostsQuotesFavs
boonhet@boonhet@sopuli.xyz
⁨1⁩h

Replying to @⁨TechAnon@lemmy.world⁩

I mean sure if your older machine’s newer GPU has 200 GB of VRAM to run a decent quant… Fucking hell just Q4 is around 100 gigs and then you’d want a decent amount of KV cache too.

0000
Open original page
BoostsQuotesFavs