[{"data":1,"prerenderedAt":369},["ShallowReactive",2],{"project-50246238":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":28,"sentiment":29,"stance_mix":30,"scored_comments":34,"aspects":35,"strip":44,"post_text":46,"what":47,"said":98,"maker_in_thread":222,"shape":233,"grid":238,"attempts":330,"same_week":331,"maker_shows":314,"maker_hidden":361,"reader_version":362,"card":363},"50246238","Hob","Open-source, on device alternative to Claude Cowork and ChatGPT","title","Show HN: Hob - Open-source, on device alternative to Claude Cowork and ChatGPT","https:\u002F\u002Fgithub.com\u002FFrigadeHQ\u002Fhob","github.com","pancomplex",1793031149,2026,177,43,343,4366,8,"notable","ai","local ai desktop app","desktop_app",[24,25,26,27],"local llm","on-device ai","open-weight models","offline ai assistant","curious",0.143,{"positive":31,"neutral":18,"mixed":32,"critical":33},4,0,3,15,{"idea":36,"docs":38,"design":39,"execution":40,"performance":42,"privacy":43},{"positive":37,"critical":32},1,{"positive":37,"critical":32},{"positive":32,"critical":32},{"positive":32,"critical":41},2,{"positive":32,"critical":37},{"positive":37,"critical":32},[45,32,32,32,32,32,32,32,32,32,32,32,32,45,32,33,32,33,37,32,37,32,37,33,32,33,33,32,32,32,45,32,32,32,45,33,32,32,33,33],5,"Hey HN, I'm Christian and this is a new OSS project I've been noodling on for the last couple of weeks. Hob is a free, MIT-licensed desktop app for Mac, Windows and Linux that works on your files the way the Claude and ChatGPT desktop apps do, except the model runs on your own computer. It uses open-weight models (currently Gemma 4 and gpt-oss). All your chats and files never leave your system. The only things that go online are web search (can be turned off in settings) and any MCP connectors you add yourself.\n\nMain features:\n\n- Automatically finds the best open-weight model for your hardware",{"problem":48,"audience":54,"pricing":62,"motivation":66,"time_to_build":71,"stack":72,"openness":95,"stage":96,"team_size":97},{"value":49,"source":50,"evidence":51},"Lets people use Claude\u002FChatGPT-style file work (reading, editing, charts, images) with models running locally for privacy","maker",[52],{"id":53},"post",{"value":55,"source":56,"evidence":57},"Privacy-conscious users who want local file-based AI assistance","inferred",[58,59],{"id":53},{"id":60,"by":11,"text":61},"50250524","Thank you! Would love to hear which ones you've used. Working with local files without anything going to the cloud was the main thing I wanted. I tried Jan, OpenWork, Chatbox and…",{"value":63,"source":50,"evidence":64},"Free",[65],{"id":53},{"value":67,"source":50,"evidence":68},"Wanted to work with local files without anything going to the cloud, which other local apps did not do well",[69,70],{"id":53},{"id":60,"by":11,"text":61},null,[73,79,83,87,91],{"value":74,"source":50,"evidence":75},"Electron",[76],{"id":77,"by":11,"text":78},"50250987","It's still pretty tricky to do this as a solo dev to keep feature parity in my experience (fun fact, even Claude's own Desktop app is made in Electron). That being said, majority…",{"value":80,"source":50,"evidence":81},"llama.cpp",[82],{"id":77,"by":11,"text":78},{"value":84,"source":50,"evidence":85},"Qwen3-VL for image recognition",[86],{"id":53},{"value":88,"source":50,"evidence":89},"Whisper for dictation",[90],{"id":53},{"value":92,"source":50,"evidence":93},"SearXNG for web search",[94],{"id":53},"open_source","unknown","solo",{"praise":99,"objections":118,"requests":143,"questions":168,"alternatives":182},[100,108],{"value":101,"source":102,"evidence":103},"Honest, handwritten, no-frills writeup and repo","commenter",[104],{"id":105,"by":106,"text":107},"50252084","mickelsen","Haven't tried it yet, but I like that you shared your process in such an honest way, and kept it no frills, as in no LLM wall-of-text readme in the repo. Wish we could see that…",{"value":109,"source":102,"evidence":110},"Open source MIT with forkable code",[111,115],{"id":112,"by":113,"text":114},"50251835","rhedak","I'm actually building something similar and thank you very much for sharing you gave me a few inspirations. I want mine to integrate seemlesly with an animated Ai companion (it's…",{"id":116,"by":11,"text":117},"50252076","Thank you. Yes please fork that's exactly why I put it out on MIT license :)",[119,129,136],{"value":120,"source":102,"evidence":121},"Local models feel too weak to be useful",[122,126],{"id":123,"by":124,"text":125},"50255506","wilg","The models are really dumb, mainly.",{"id":127,"by":124,"text":128},"50253770","I did not enjoy that there was no way for it to clean up the 18GB of models it installed after I determined it was not really usable for any real work.",{"value":130,"source":102,"evidence":131},"Installs large model downloads with no cleanup option",[132,133],{"id":127,"by":124,"text":128},{"id":134,"by":11,"text":135},"50254415","Sorry about that. I’ll make sure it cleans up after itself in the next update. What was the main issue you ran into?",{"value":137,"source":102,"evidence":138},"Electron\u002FJS instead of native apps",[139],{"id":140,"by":141,"text":142},"50250891","_aavaa_","Why does it have to be JS? With all this AI, why can't we ship native apps.",[144,151,161],{"value":145,"source":102,"evidence":146},"Let users change the model installation directory",[147],{"id":148,"by":149,"text":150},"50250615","laughing_snyder","Looks great. One thing I am missing is the ability to change the installation directory for the models. For space reasons, I want them to be on a different partition.",{"value":152,"source":102,"evidence":153},"Support local network LLM servers like Ollama",[154,158],{"id":155,"by":156,"text":157},"50252827","atmanactive","Can we configure it to use local models but in LAN? Like Ollama or Strata.",{"id":159,"by":156,"text":160},"50253983","Yes.",{"value":162,"source":102,"evidence":163},"Benchmarks comparing models",[164],{"id":165,"by":166,"text":167},"50265024","iqbal1980","Thank you so much, that's helpful. Do you offer any benchmarks comparing with others?",[169,176],{"value":170,"source":102,"evidence":171},"Which hardware runs which model",[172],{"id":173,"by":174,"text":175},"50251374","eecks","> Automatically finds the best open-weight model for your hardware on first install Do you have examples of what hardware runs what models?",{"value":177,"source":102,"evidence":178},"Which model suits a 16GB M5 Mac for email and marketing",[179],{"id":180,"by":166,"text":181},"50258835","Interesting, thanks for sharing this. I have some questions? I have a Macbook M5 with 16GB unified memory. What Model would you suggest I use with this. I want something that…",[183,194,201,211,218],{"value":184,"source":102,"evidence":185},"Osaurus",[186,190],{"id":187,"by":188,"text":189},"50250735","universenz","Definitely check out Osaurus for macOS and Windows. Heavily privacy focused with e2ee iOS and iPadOS tethers.",{"id":191,"by":192,"text":193},"50265580","dofm","No worries. And yes, absolutely. Even the tiniest opportunity to poke around with stuff changes my perspective dramatically. I've picked up lots of stuff this way — just being…",{"value":195,"source":102,"evidence":196},"Open WebUI",[197],{"id":198,"by":199,"text":200},"50251092","prophesi","Have you tried Open WebUI? It doesn't determine which model works best for your hardware, but it otherwise has a very large feature set with a large community, and runs offline…",{"value":202,"source":102,"evidence":203},"LM Studio",[204,208],{"id":205,"by":206,"text":207},"50250464","vorticalbox","Did you happen to try bionic? https:\u002F\u002Flmstudio.ai\u002F",{"id":209,"by":192,"text":210},"50253704","Shouldn’t be embarrassing to admit. It’s a huge new subject area and it’s completely exhausting to learn. I still feel like a total novice. I have an M1 Max (more memory…",{"value":212,"source":102,"evidence":213},"Orca",[214],{"id":215,"by":216,"text":217},"50251141","ajvs","Orca works great.",{"value":219,"source":50,"evidence":220},"Jan, OpenWork, Chatbox, Lobe",[221],{"id":60,"by":11,"text":61},{"comments":223,"answered":224,"key":225},11,6,[226,227,230],{"id":60,"text":61},{"id":228,"text":229},"50252181","thank you! Haha I always make it a point these days do handwrite these things. I think everyone is tired of overly verbose perfectly written AI text.",{"id":231,"text":232},"50250637","Good call. Are you on PC\u002FMac\u002FLinux?",{"commenters":234,"max_depth":235,"maker_replies":223,"first_comment_mins":236,"lifespan_hours":237},22,7,254,66,[239,243,245,247,250,251,254,255,259,261,263,264,265,268,270,272,273,275,276,278,280,282,283,284,286,288,290,291,294,296,298,300,301,305,307,310,312,317,320,322,324,326,329],[240,32,45,241,242,32],"50250424","project\u002Fpositive\u002Fidea","jrm4",[60,32,32,244,11,37],"maker\u002Fneutral\u002Fnone",[187,32,32,246,188,32],"other\u002Fneutral\u002Fnone",[248,32,32,246,249,32],"50250856","fmxexpress",[198,32,32,246,199,32],[252,32,32,246,253,32],"50253609","dfltr",[209,32,32,246,192,32],[256,32,32,257,258,32],"50261259","commenter\u002Fneutral\u002Fnone","felciano",[260,32,32,246,192,32],"50262331",[262,32,32,257,253,32],"50264016",[191,32,32,257,192,32],[215,32,32,246,216,32],[266,32,32,246,267,32],"50251534","l3x4ur1n",[269,32,32,246,242,32],"50252072",[105,37,45,271,106,32],"maker\u002Fpositive\u002Fdocs",[228,37,32,257,11,37],[148,41,33,274,149,32],"project\u002Fneutral\u002Fdesign",[231,41,32,244,11,37],[277,41,33,274,149,32],"50250701",[140,33,37,279,141,32],"project\u002Fcritical\u002Fexecution",[77,33,32,281,11,37],"maker\u002Fneutral\u002Fexecution",[127,31,37,279,124,32],[134,31,32,281,11,37],[123,31,37,285,124,32],"project\u002Fcritical\u002Fperformance",[173,45,33,287,174,32],"project\u002Fneutral\u002Fperformance",[289,45,32,287,11,37],"50252050",[180,224,33,287,166,32],[292,224,32,293,11,37],"50262305","maker\u002Fneutral\u002Fperformance",[165,224,33,295,166,32],"maker\u002Fneutral\u002Fdocs",[297,224,32,295,11,37],"50265771",[112,235,32,299,113,32],"commenter\u002Fneutral\u002Fidea",[116,235,32,257,11,37],[302,18,45,303,304,32],"50256093","project\u002Fpositive\u002Fprivacy","arkdezin",[205,306,32,246,206,32],9,[308,306,32,309,149,32],"50250598","other\u002Fneutral\u002Fprivacy",[311,306,32,309,188,32],"50250800",[313,314,45,315,316,32],"50286975",10,"project\u002Fpositive\u002Fnone","artsaraiva",[318,223,33,244,319,32],"50253249","ourcat",[321,223,32,244,11,37],"50253312",[323,223,32,246,192,32],"50253637",[155,325,33,274,156,32],12,[327,325,32,328,11,37],"50252873","maker\u002Fneutral\u002Fdesign",[159,325,33,328,156,32],[],[332,337,342,347,352,356],{"id":333,"name":334,"points":335,"category":336},"50224916","Steal My Bitcoin",625,"learning_science",{"id":338,"name":339,"points":340,"category":341},"50238708","TitleArt",409,"creative",{"id":343,"name":344,"points":345,"category":346},"50228115","keybytes",320,"hardware",{"id":348,"name":349,"points":350,"category":351},"50195488","OnKey",237,"consumer_social",{"id":353,"name":354,"points":355,"category":346},"50260753","Marelleo",194,{"id":357,"name":358,"points":359,"category":360},"50232312","Big Whale Hunting workforce visualisation",178,"finance_crypto",false,"v2",{"id":364,"name":5,"line":6,"ai":361,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":365,"top":18,"kw":366,"maker":11,"domain":10,"tier":19,"attempts":32,"strip":367,"url":9,"mix":368,"praise":101,"objection":120,"sub":21},50246238,0.14,[24,25,26,27],"5000000000000503031010130330005000530033",[31,18,32,33],1793991170567]