[{"data":1,"prerenderedAt":263},["ShallowReactive",2],{"project-38295179":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":29,"sentiment":30,"stance_mix":31,"scored_comments":35,"aspects":36,"strip":44,"post_text":45,"what":46,"said":74,"maker_in_thread":159,"shape":170,"grid":174,"attempts":225,"same_week":226,"maker_shows":35,"maker_hidden":255,"reader_version":256,"card":257},"38295179","Tiny LLMs","Browser-based private AI models for a wide array of tasks","title","Show HN: Tiny LLMs – Browser-based private AI models for a wide array of tasks","https:\u002F\u002Ftinyllms.vercel.app\u002Fdashboard","tinyllms.vercel.app","bilater",1700167414,2023,142,23,448,3109,15,"notable","ai","browser-based ai models","web_app",[24,25,26,27,28],"browser ai","speech to text","webgpu","transformers.js","private ai","roasted",-0.189,{"positive":32,"neutral":32,"mixed":33,"critical":34},3,1,7,14,{"execution":37,"performance":40,"idea":41,"docs":43},{"positive":38,"critical":39},0,2,{"positive":39,"critical":38},{"positive":33,"critical":42},5,{"positive":38,"critical":38},[32,42,33,38,33,38,32,32,33,42,38,38,38,33,42,33,38,33,33,32,38,38,38],null,{"problem":47,"audience":45,"pricing":45,"motivation":45,"time_to_build":45,"stack":57,"openness":72,"stage":73,"team_size":72},{"value":48,"source":49,"evidence":50},"Running AI models privately in the browser without a server","inferred",[51,53],{"id":52},"post",{"id":54,"by":55,"text":56},"38297332","anigbrowl","You should have some text explaining what is going and how they are running locally on the GPU - otherwise everyone sharing the link will need to write an explanation of why…",[58,65],{"value":27,"source":59,"evidence":60},"commenter",[61],{"id":62,"by":63,"text":64},"38296267","sunshadow","Original project: https:\u002F\u002Fgithub.com\u002Fxenova\u002Ftransformers.js",{"value":66,"source":59,"evidence":67},"Whisper (via whisper-web port)",[68],{"id":69,"by":70,"text":71},"38296151","albertzeyer","The name \"Tiny LLMs\" does not make sense. LLM stands for large language model. So tiny is kind of the opposite of large. Large usually means >=7B parameters (2023). If it's less,…","unknown","launched",{"praise":75,"objections":87,"requests":120,"questions":133,"alternatives":148},[76],{"value":77,"source":59,"evidence":78},"Speech-to-text in browser is impressive and works well, even on Firefox on Mac",[79,83],{"id":80,"by":81,"text":82},"38296233","jackconsidine","The speech recognition tool is impressive! I've tried many different attempts at browser-local speech-to-text from Ermine.ai to Google's Tensorflow.js speech toolkit (recognition…",{"id":84,"by":85,"text":86},"38296636","raybb","Love that the speech to text even works in FF on Mac, and rather well I'd say.",[88,108,116],{"value":89,"source":59,"evidence":90},"Name is misleading; the tools are not LLMs",[91,95,99,103,107],{"id":92,"by":93,"text":94},"38296083","TOMDM","Which one of these is an LLM? The youtube summary one maybe? Doesn't seem to run in the browser.",{"id":96,"by":97,"text":98},"38296120","ilaksh","None of these are LLMs. LLM means Large Language Model (for text generation, like ChatGPT).",{"id":100,"by":101,"text":102},"38296155","politelemon","Tiny Models would have been a better name. LLMs are a specific type of model that generate text. Or Tiny GenAI to capitalize on the marketing names?",{"id":104,"by":105,"text":106},"38311192","gsuuon","This is cool but named pretty poorly. It doesn't seem to have anything to do with LLM's - TinyAIs would've been perfect.",{"id":69,"by":70,"text":71},{"value":109,"source":59,"evidence":110},"Tools shown do not appear to be language models or run in browser",[111,112],{"id":92,"by":93,"text":94},{"id":113,"by":114,"text":115},"38303393","supermatt","These aren't LLMs...",{"value":117,"source":59,"evidence":118},"Lacks explanation of what the site does and how it runs",[119],{"id":54,"by":55,"text":56},[121,128],{"value":122,"source":59,"evidence":123},"Add a roughly 1B parameter LLM for the browser",[124],{"id":125,"by":126,"text":127},"38296139","senseiV","Nice, whats the text to video model? Also, you could try to go for a 1b llm for the browser, would fit.",{"value":129,"source":59,"evidence":130},"Add more explanation and model details",[131,132],{"id":54,"by":55,"text":56},{"id":69,"by":70,"text":71},[134,144],{"value":135,"source":59,"evidence":136},"What is Mistral 1B and is it supported",[137,141],{"id":138,"by":139,"text":140},"38296250","woadwarrior01","What’s Mistral 1B? I only know of Mistral 7B.",{"id":142,"by":139,"text":143},"38304545","The announcement seems somewhat disingenuous. The PR[1] found from their release notes[2] seems to contain only boilerplate and no real support for Mistral models or their…",{"value":145,"source":59,"evidence":146},"Which tool on the site is an LLM",[147],{"id":92,"by":93,"text":94},[149,155],{"value":150,"source":59,"evidence":151},"WebLLM browser chat demo",[152],{"id":153,"by":97,"text":154},"38296126","Here is an actual LLM that runs in the browser: https:\u002F\u002Fwebllm.mlc.ai\u002F#chat-demo I cannot make this work on Chrome in Linux. I even tried with…",{"value":156,"source":59,"evidence":157},"Ermine.ai and Google TensorFlow.js speech toolkit",[158],{"id":80,"by":81,"text":82},{"comments":32,"answered":32,"key":160},[161,164,167],{"id":162,"text":163},"38296098","The other tools section is just extra helpful links to other stuff I've done.",{"id":165,"text":166},"38296150","Good idea! Mistral 1B is supported by transformers.js now so I can add it! The Text to Video model is a pipeline Text to Speech + FFMPEG Magic to stitch together a video.",{"id":168,"text":169},"38296314","I think it might be a \u003C 1B fine tuned model. Read about it in this release https:\u002F\u002Ftwitter.com\u002Fxenovacom\u002Fstatus\u002F1722661501180256311",{"commenters":171,"max_depth":42,"maker_replies":32,"first_comment_mins":172,"lifespan_hours":173},18,68,34,[175,177,179,181,183,185,189,190,193,194,197,199,201,202,204,205,207,210,212,214,217,220,223],[153,38,32,176,97,38],"project\u002Fneutral\u002Fexecution",[80,33,42,178,81,38],"project\u002Fpositive\u002Fperformance",[92,39,33,180,93,38],"project\u002Fcritical\u002Fexecution",[162,39,38,182,11,33],"project\u002Fneutral\u002Fnone",[96,39,33,184,97,38],"project\u002Fcritical\u002Fidea",[186,39,38,187,188,38],"38296470","commenter\u002Fcritical\u002Fidea","kordlessagain",[62,32,32,182,63,38],[54,191,32,192,55,38],4,"maker\u002Fmixed\u002Fdocs",[100,42,33,184,101,38],[125,195,42,196,126,38],6,"project\u002Fpositive\u002Fidea",[165,195,38,198,11,33],"project\u002Fneutral\u002Fidea",[138,195,38,200,139,38],"commenter\u002Fneutral\u002Fnone",[168,195,38,198,11,33],[142,195,33,203,139,38],"maker\u002Fcritical\u002Fexecution",[84,34,42,178,85,38],[104,206,33,184,105,38],8,[208,206,38,200,209,38],"38316442","hilti",[113,211,33,184,114,38],9,[69,213,33,184,70,38],10,[215,213,32,198,216,38],"38296191","unshavedyak",[218,213,38,200,219,38],"38296380","andy99",[221,213,38,200,222,38],"38296521","zamadatix",[224,213,38,200,70,38],"38296709",[],[227,232,237,241,246,250],{"id":228,"name":229,"points":230,"category":231},"38327017","Ad Accelerator",713,"consumer_social",{"id":233,"name":234,"points":235,"category":236},"38316936","UneeBee",459,"learning_science",{"id":238,"name":239,"points":240,"category":236},"38365638","Pickcode",420,{"id":242,"name":243,"points":244,"category":245},"38285228","screenshot-to-code",358,"dev_tools",{"id":247,"name":248,"points":249,"category":20},"38368570","Neum AI",155,{"id":251,"name":252,"points":253,"category":254},"38360776","Stella Nera",154,"hardware",false,"v2",{"id":258,"name":5,"line":6,"ai":255,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":259,"top":18,"kw":260,"maker":11,"domain":10,"tier":19,"attempts":38,"strip":261,"url":9,"mix":262,"praise":77,"objection":89,"sub":21},38295179,-0.19,[24,25,26,27,28],"35101033150001510113000",[32,32,33,34],1794036644806]