mapumbaa
- 97 Posts
- 321 Comments
mapumbaa@lemmy.zipOPto Sweden@lemmy.world•Lawen Redar spökar runt i Ulf Kristerssons huvud2·16 days agoBra avsnitt.
I wouldn’t worry too much about accidents. But people will get upset if you disturb the hunt. At least during the first couple of weeks.
mapumbaa@lemmy.zipto AMD@lemmy.zip•Like Intel before it, AMD blames motherboard makers for burnt-out CPUsEnglish5·23 days agoMakes sense. Considering the low quality of the software that comes out of ASRock and alike, I’m surprised anything works at all.
mapumbaa@lemmy.zipto Hardware@lemmy.world•Softbank reportedly considered buying Intel's foundry division outright before investing $2 billion into the company as equityEnglish1·28 days agoThe US should nationalize the foundry. Make it a federal service. Only thing that makes sense.
mapumbaa@lemmy.zipto Intel@lemmy.zip•Intel is spinning off its Network and Edge groupEnglish1·29 days agoThis might actually be a good investment for Ericsson, unlike all the cloud BS they’ve been throwing money at for the past 15 years.
The Intel foundries should be nationalized. It does not make sense to have private players when there is no functional market. The EU should start/buy its own chip manufacturing business as well.
mapumbaa@lemmy.zipOPto Sweden@lemmy.world•Tips: Nya Biblio-appen har inbyggt stöd för TTS av e-böcker1·30 days agoFunktionen verkar dock ännu vara ganska buggig och undermålig. Tyvärr.
mapumbaa@lemmy.zipOPto Sweden@lemmy.world•Tips: Nya Biblio-appen har inbyggt stöd för TTS av e-böcker1·30 days agoText-to-speech (uppläsning). Borde skrivit på svenska.
mapumbaa@lemmy.zipto Hardware@lemmy.world•The U.S. Is Discussing Taking a Stake in Intel [WSJ]English2·1 month agoGood. What I’ve been advocating. Since there is no functional market, it doesn’t not make sense to have private players.
mapumbaa@lemmy.zipOPto LocalLLaMA@sh.itjust.works•HP Z2 Mini G1a Review: Running GPT-OSS 120B Without a Discrete GPUEnglish21·1 month agoThere is nothing “optimized” that will get you better inference performance of medium/large models at $2000.
mapumbaa@lemmy.zipOPto LocalLLaMA@sh.itjust.works•HP Z2 Mini G1a Review: Running GPT-OSS 120B Without a Discrete GPUEnglish21·1 month agoI promise. It’s not possible. But things change quickly of course.
(Unless you’re lucky/pro and get your hands on some super cheap used high end hardware…)
mapumbaa@lemmy.zipOPto LocalLLaMA@sh.itjust.works•HP Z2 Mini G1a Review: Running GPT-OSS 120B Without a Discrete GPUEnglish2·1 month agoIt will probably kick the ass of that desktop. $2000 won’t get you far with a conventional build.
mapumbaa@lemmy.zipOPto LocalLLaMA@sh.itjust.works•GPT-OSS 20B and 120B Models on AMD Ryzen AI ProcessorsEnglish2·1 month agoThat you have to wait until Christmas when you can get a HP mini today: https://www.storagereview.com/review/hp-z2-mini-g1a-review-running-gpt-oss-120b-without-a-discrete-gpu
mapumbaa@lemmy.zipto Free Open-Source Artificial Intelligence@lemmy.world•Do you use models developed in China?English1·1 month agoI believe the full size DeepSeek-R1 require about 1200 GB of VRAM. But there are many configurations that require much less. Quantization, MoE and other hacks. I don’t have much experience with MoE, however I find that quantization tend to decrease performance significantly. At least with models from Mistral.
mapumbaa@lemmy.zipto A Boring Dystopia@lemmy.world•Peter Thiel’s bestie going mask offEnglish9·2 months agoIs there any point in doing that? All these guys are openly racist, anti-democracy and anti-america. Enemies of the state by definition (a state which is almost entierly under their control by now). We’re way beyond calling out on racism.
mapumbaa@lemmy.zipto Privacy@lemmy.ml•[OC] What People Think Privacy/Security IsEnglish5·2 months agoWell, usually not. Unfortunately.
Min bild är att SVT har blivit helt kastrerat. Det enda de bryr sig om är att inte kliva någon folkmordsrunkare på tårna. Klarar de av att utföra någon form av seriös journalistik nuförtiden?