ikt@aussie.zoneEnglish · 2 days agoETH Zurich and EPFL will release a large language model (LLM) developed on public infrastructure. Trained on the “Alps” supercomputer at the Swiss National Supercomputing Centre (CSCS)plus-squareethz.chexternal-linkmessage-square1fedilinkarrow-up128arrow-down10
arrow-up128arrow-down1external-linkETH Zurich and EPFL will release a large language model (LLM) developed on public infrastructure. Trained on the “Alps” supercomputer at the Swiss National Supercomputing Centre (CSCS)plus-squareethz.chikt@aussie.zoneEnglish · 2 days agomessage-square1fedilink
SmokeyDope@lemmy.worldMEnglish · edit-24 days agoHow to calculate cost-per-tokens output of local model compared to enterprise model API accessplus-squarelemmy.worldimagemessage-square12fedilinkarrow-up130arrow-down11
arrow-up129arrow-down1imageHow to calculate cost-per-tokens output of local model compared to enterprise model API accessplus-squarelemmy.worldSmokeyDope@lemmy.worldMEnglish · edit-24 days agomessage-square12fedilink
brokenlcd@feddit.itEnglish · edit-24 days agoneed help understanding if this setup is even feasible.plus-squaremessage-squaremessage-square18fedilinkarrow-up111arrow-down12
arrow-up19arrow-down1message-squareneed help understanding if this setup is even feasible.plus-squarebrokenlcd@feddit.itEnglish · edit-24 days agomessage-square18fedilink
robber@lemmy.mlEnglish · 7 days agoNew, promising MoE model "Hunyuan" by Tencentplus-squarehuggingface.coexternal-linkmessage-square1fedilinkarrow-up115arrow-down11
arrow-up114arrow-down1external-linkNew, promising MoE model "Hunyuan" by Tencentplus-squarehuggingface.corobber@lemmy.mlEnglish · 7 days agomessage-square1fedilink
wise_pancake@lemmy.caEnglish · 17 days agoAre you using any MCP servers? If so, what?message-squaremessage-square8fedilinkarrow-up131arrow-down10
arrow-up131arrow-down1message-squareAre you using any MCP servers? If so, what?wise_pancake@lemmy.caEnglish · 17 days agomessage-square8fedilink
SmokeyDope@lemmy.worldMEnglish · edit-217 days agoHomelab upgrade WIPplus-squarelemmy.worldimagemessage-square4fedilinkarrow-up127arrow-down11
arrow-up126arrow-down1imageHomelab upgrade WIPplus-squarelemmy.worldSmokeyDope@lemmy.worldMEnglish · edit-217 days agomessage-square4fedilink
robber@lemmy.mlEnglish · 19 days agoDo you quantize models yourself?plus-squarelemmy.mlimagemessage-square7fedilinkarrow-up124arrow-down10
arrow-up124arrow-down1imageDo you quantize models yourself?plus-squarelemmy.mlrobber@lemmy.mlEnglish · 19 days agomessage-square7fedilink
Womble@lemmy.worldEnglish · edit-220 days agoAny recomendations for a text editor with good AI integration (not code editor)plus-squaremessage-squaremessage-square11fedilinkarrow-up120arrow-down13
arrow-up117arrow-down1message-squareAny recomendations for a text editor with good AI integration (not code editor)plus-squareWomble@lemmy.worldEnglish · edit-220 days agomessage-square11fedilink
splendoruranium@infosec.pubEnglish · edit-21 month agoLocal Voiceover/Audiobook generationplus-squaremessage-squaremessage-square6fedilinkarrow-up127arrow-down10
arrow-up127arrow-down1message-squareLocal Voiceover/Audiobook generationplus-squaresplendoruranium@infosec.pubEnglish · edit-21 month agomessage-square6fedilink
SmokeyDope@lemmy.worldMEnglish · edit-21 month agoMistralAI releases Magistral, their first official reasoning models. magistral small 2506 released under apache 2.0 license!plus-squaremistral.aiexternal-linkmessage-square4fedilinkarrow-up137arrow-down13
arrow-up134arrow-down1external-linkMistralAI releases Magistral, their first official reasoning models. magistral small 2506 released under apache 2.0 license!plus-squaremistral.aiSmokeyDope@lemmy.worldMEnglish · edit-21 month agomessage-square4fedilink
10001110101@lemm.eeEnglish · 1 month agoCurrent best local models for tool use?plus-squaremessage-squaremessage-square8fedilinkarrow-up114arrow-down11
arrow-up113arrow-down1message-squareCurrent best local models for tool use?plus-square10001110101@lemm.eeEnglish · 1 month agomessage-square8fedilink
pebbles@sh.itjust.worksEnglish · 1 month agoI'm excited for dots.llm (142BA14B)!plus-squaregithub.comexternal-linkmessage-square3fedilinkarrow-up112arrow-down10
arrow-up112arrow-down1external-linkI'm excited for dots.llm (142BA14B)!plus-squaregithub.compebbles@sh.itjust.worksEnglish · 1 month agomessage-square3fedilink
HumanPerson@sh.itjust.worksEnglish · 1 month agoWhat to Integrate My AI Withplus-squaremessage-squaremessage-square6fedilinkarrow-up119arrow-down13
arrow-up116arrow-down1message-squareWhat to Integrate My AI Withplus-squareHumanPerson@sh.itjust.worksEnglish · 1 month agomessage-square6fedilink
SmokeyDope@lemmy.worldMEnglish · edit-21 month agoUpdated guidelines for c/LocaLLama (new rules)plus-squaremessage-squaremessage-square8fedilinkarrow-up128arrow-down10
arrow-up128arrow-down1message-squareUpdated guidelines for c/LocaLLama (new rules)plus-squareSmokeyDope@lemmy.worldMEnglish · edit-21 month agomessage-square8fedilink
ikt@aussie.zoneEnglish · 1 month agoAMD’s Untether AI Deal - Bad Signs for GPU-Driven AI trainingplus-squaresemiconductorsinsight.comexternal-linkmessage-square15fedilinkarrow-up134arrow-down10
arrow-up134arrow-down1external-linkAMD’s Untether AI Deal - Bad Signs for GPU-Driven AI trainingplus-squaresemiconductorsinsight.comikt@aussie.zoneEnglish · 1 month agomessage-square15fedilink
ikt@aussie.zoneEnglish · edit-21 month agoMy AI Skeptic Friends Are All Nutsplus-squarenews.ycombinator.comexternal-linkmessage-square13fedilinkarrow-up12arrow-down12
arrow-up10arrow-down1external-linkMy AI Skeptic Friends Are All Nutsplus-squarenews.ycombinator.comikt@aussie.zoneEnglish · edit-21 month agomessage-square13fedilink
WalnutLum@lemmy.mlEnglish · 1 month agoDespite record gaming revenues, Nvidia reportedly plans to cut RTX 50 series production to allocate it to new AI hardwareplus-squareoverclock3d.netexternal-linkmessage-square2fedilinkarrow-up11arrow-down10
arrow-up11arrow-down1external-linkDespite record gaming revenues, Nvidia reportedly plans to cut RTX 50 series production to allocate it to new AI hardwareplus-squareoverclock3d.netWalnutLum@lemmy.mlEnglish · 1 month agomessage-square2fedilink
ThreeJawedChuck@sh.itjust.worksEnglish · 1 month agoNoob experience using local LLM as a D&D style DM.plus-squaremessage-squaremessage-square6fedilinkarrow-up10arrow-down10
arrow-up10arrow-down1message-squareNoob experience using local LLM as a D&D style DM.plus-squareThreeJawedChuck@sh.itjust.worksEnglish · 1 month agomessage-square6fedilink
ThreeJawedChuck@sh.itjust.worksEnglish · 1 month agoDon't overlook llama.cpp's rpc-server feature.plus-squaremessage-squaremessage-square4fedilinkarrow-up11arrow-down10
arrow-up11arrow-down1message-squareDon't overlook llama.cpp's rpc-server feature.plus-squareThreeJawedChuck@sh.itjust.worksEnglish · 1 month agomessage-square4fedilink
thickertoofan@lemm.eeEnglish · 1 month agoLLMs and their efficiency, can they really replace humans?plus-squarearxiv.orgexternal-linkmessage-square1fedilinkarrow-up11arrow-down10
arrow-up11arrow-down1external-linkLLMs and their efficiency, can they really replace humans?plus-squarearxiv.orgthickertoofan@lemm.eeEnglish · 1 month agomessage-square1fedilink
