[{"data":1,"prerenderedAt":265},["ShallowReactive",2],{"project-38730143":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":9,"maker":10,"posted_at":11,"year":12,"points":13,"comments":14,"rank_in_year":15,"shown_in_year":16,"top":17,"tier":18,"category":19,"subcategory":20,"format":21,"keywords":22,"reception":26,"sentiment":27,"stance_mix":28,"scored_comments":33,"aspects":34,"strip":41,"post_text":42,"what":43,"said":66,"maker_in_thread":163,"shape":166,"grid":169,"attempts":227,"same_week":228,"maker_shows":31,"maker_hidden":256,"reader_version":257,"card":258},"38730143","Emu2","A Gemini-like open-source 37B Multimodal Model","title","Show HN: Emu2 – A Gemini-like open-source 37B Multimodal Model",null,"BAAIBeijing",1703211605,2023,153,27,406,3109,14,"notable","ai","multimodal model","other",[20,23,24,25],"open source llm","image generation","visual understanding","curious",0.094,{"positive":29,"neutral":30,"mixed":31,"critical":32},5,16,1,4,26,{"docs":35,"idea":37,"performance":39},{"positive":36,"critical":36},0,{"positive":38,"critical":36},2,{"positive":31,"critical":40},3,[40,29,29,40,36,40,40,40,31,40,40,29,40,31,40,40,40,40,31,40,29,29,40,40,40,31,40],"Hello HN, I'm excited to introduce Emu2, the latest generative multimodal model developed by the Beijing Academy of Artificial Intelligence (BAAI). Emu2 is an open-source initiative that reflects BAAI's commitment to fostering open, secure, and responsible AI research. It's designed to enhance AI's proficiency in handling tasks across various modalities with minimal examples and straightforward instructions.\n\nEmu2 has demonstrated superior performance over other large-scale models like Flamingo-80B in few-shot multimodal understanding tasks. It serves as a versatile base model for developers, ",{"problem":9,"audience":44,"pricing":9,"motivation":50,"time_to_build":9,"stack":54,"openness":63,"stage":64,"team_size":65},{"value":45,"source":46,"evidence":47},"Developers building multimodal applications","maker",[48],{"id":49},"post",{"value":51,"source":46,"evidence":52},"Foster open, secure and responsible AI research",[53],{"id":49},[55],{"value":56,"source":57,"evidence":58},"Llama 1 (model initialized from)","commenter",[59],{"id":60,"by":61,"text":62},"38760308","F_AnZzz","To be more specified, our codes follows the license of Apache 2.0 and the model weights follows the license of Llama 1 since our model is initialized from Llama1","open_source","launched","company",{"praise":67,"objections":86,"requests":117,"questions":133,"alternatives":148},[68,79],{"value":69,"source":57,"evidence":70},"Open-sourcing a capable multimodal model is welcome",[71,75],{"id":72,"by":73,"text":74},"38730959","opdahl","I’m always in favor of more open-source models, and especially multimodal ones. I feel that the end goal for these models will be multimodal so it’s important that open-source…",{"id":76,"by":77,"text":78},"38731751","hao_wen","Wow! It is really amazing that such a capable large-scale multimodal model is open-sourced. Thank you for sharing this.",{"value":80,"source":57,"evidence":81},"Performs well in tests",[82],{"id":83,"by":84,"text":85},"38731431","Vt71fcAqt7","This is really cool, thanks for sharing this! I tried a few tests and it did pretty well. But it translated 热闹 as \"hot and noisy\" with this prompt: Give some example sentences…",[87,96,110],{"value":88,"source":57,"evidence":89},"Primarily trained on English, weak on other languages",[90,93],{"id":91,"by":61,"text":92},"38732792","Emu2 is primarily trained on the English corpus. Hence, it may not perform well as expected when dealing with tasks like translation.",{"id":94,"by":61,"text":95},"38732768","Emu2 is primarily trained on the English corpus, which may result in suboptimal performance when dealing with inputs in other languages.",{"value":97,"source":57,"evidence":98},"Licensing was unclear and model weights under GPLv3 at first",[99,103,106],{"id":100,"by":101,"text":102},"38730760","krasin","Thank you for posting this announce! I tried to find under which license this work is available, but it does not appear to be stated anywhere. Without the license being clear,…",{"id":104,"by":10,"text":105},"38731111","The current model weights are distributed under the GPLv3 license. The authors are in the process of releasing a new series of Emu2 models that will be accompanied by a more…",{"id":107,"by":108,"text":109},"38732526","zarzavat","Applying GPL to model weights is very odd indeed.",{"value":111,"source":57,"evidence":112},"Hugging Face demo has long queue times",[113],{"id":114,"by":115,"text":116},"38731382","isaacfung","The examples look very promising but the queue time of the huggingface demo is a bit long I actually got a timeout error :(",[118,126],{"value":119,"source":57,"evidence":120},"Clarify the license",[121,122],{"id":100,"by":101,"text":102},{"id":123,"by":124,"text":125},"38730780","Zuiii","The first question I ask myself when I see any new model: What's the license? Edit: Also couldn't find a license.",{"value":127,"source":57,"evidence":128},"Release the chat template",[129],{"id":130,"by":131,"text":132},"38733674","xena","What is the chat template? I can't find it in the model card or model details.",[134,141],{"value":135,"source":57,"evidence":136},"How does it compare to ChatGPT or LLaMA on Chinese translation?",[137],{"id":138,"by":139,"text":140},"38732250","getwiththeprog","Does Emu2 have better Chinese content and translation than say ChatGPT or LLAMA?",{"value":142,"source":57,"evidence":143},"Is it similar to GPT-4-Vision?",[144],{"id":145,"by":146,"text":147},"38737094","moochee","Is it similar to OpenAI GPT-4-Vision model?",[149,153,159],{"value":150,"source":46,"evidence":151},"Flamingo-80B",[152],{"id":49},{"value":154,"source":57,"evidence":155},"LLaVA",[156],{"id":157,"by":115,"text":158},"38732284","Wow, the other link is so much faster. Almost the same speed as the much weaker llava demo on huggingface.",{"value":160,"source":57,"evidence":161},"ChatGPT",[162],{"id":138,"by":139,"text":140},{"comments":31,"answered":36,"key":164},[165],{"id":104,"text":105},{"commenters":17,"max_depth":32,"maker_replies":31,"first_comment_mins":167,"lifespan_hours":168},93,96,[170,172,175,177,178,179,183,185,187,189,191,193,195,197,199,200,202,204,208,210,213,215,217,220,222,224,225],[100,36,40,171,101,36],"project\u002Fneutral\u002Fnone",[173,36,29,174,61,36],"38733544","maker\u002Fpositive\u002Fnone",[176,36,29,174,101,36],"38738860",[60,36,40,171,61,36],[104,31,36,171,10,31],[180,31,40,181,182,36],"38731884","maker\u002Fneutral\u002Fdocs","da4id",[184,31,40,181,61,36],"38733552",[186,31,40,171,61,36],"38760312",[107,31,31,188,108,36],"project\u002Fcritical\u002Fnone",[190,31,40,181,61,36],"38733550",[192,31,40,171,61,36],"38760309",[72,38,29,194,73,36],"project\u002Fpositive\u002Fidea",[83,40,40,196,84,36],"project\u002Fmixed\u002Fperformance",[91,40,31,198,61,36],"project\u002Fcritical\u002Fperformance",[123,32,40,171,124,36],[201,32,40,181,61,36],"38733547",[203,32,40,171,61,36],"38760315",[205,29,40,206,207,36],"38732022","project\u002Fneutral\u002Fidea","vittore",[114,209,31,198,115,36],6,[211,209,40,212,61,36],"38731823","project\u002Fneutral\u002Fperformance",[157,209,29,214,115,36],"project\u002Fpositive\u002Fperformance",[76,216,29,194,77,36],7,[130,218,40,219,131,36],8,"project\u002Fneutral\u002Fdocs",[221,218,40,181,61,36],"38768285",[138,223,40,212,139,36],9,[94,223,31,198,61,36],[145,226,40,171,146,36],10,[],[229,234,238,242,247,252],{"id":230,"name":231,"points":232,"category":233},"38733968","Heynote",1063,"productivity",{"id":235,"name":236,"points":237,"category":233},"38787892","Rem",555,{"id":239,"name":240,"points":241,"category":233},"38747710","Storyteller",338,{"id":243,"name":244,"points":245,"category":246},"38772382","Whataaabout.com",235,"consumer_social",{"id":248,"name":249,"points":250,"category":251},"38717114","Talk2Arxiv",194,"learning_science",{"id":253,"name":254,"points":255,"category":233},"38753911","Weeks of Your",145,false,"v2",{"id":259,"name":5,"line":6,"ai":256,"cat":19,"year":12,"points":13,"comments":14,"rec":26,"s":260,"top":17,"kw":261,"maker":10,"domain":262,"tier":18,"attempts":36,"strip":263,"url":-1,"mix":264,"praise":69,"objection":88,"sub":20},38730143,0.09,[20,23,24,25],"","355303331335313333135533313",[29,30,31,32],1794036646529]