AI专题：从模型视角看端侧AI：模型技术持续演进交互体验有望升级

西南证券股份有...07-30

基础的构建：模型实现高效压缩是端侧AI的第一步。模型尺寸变小、同时具备较好性能，是端侧AI的前提。目前，在10B参数规模以下的模型中，7B尺寸占据主流，3B及以下小模型仍在探索，部分小模型性能正逐步接近更大参数模型，如谷歌Gemini-Nano模型在部分测试基准上接近Gemini-Pro、Meta Llama-3-8B模型表现可与Llama-2-70B匹敌。模型厂商为兼顾模型尺寸与性能，在算法优化...

网页链接

免责声明：本文观点仅代表作者个人观点，不构成本平台的投资建议，本平台不对文章信息准确性、完整性和及时性做出任何保证，亦不对因使用或信赖文章信息引发的任何损失承担责任。

精彩评论

我们需要你的真知灼见来填补这片空白

发表看法

{"i18n":{"language":"zh_CN"},"isChannel":false,"data":{"share":"https://www.laohu8.com/m/news/2455803089?lang=zh_CN&edition=full","thumbnail":"","is_english":false,"pubTime":"2024-07-30 00:00","share_image_url":"https://static.laohu8.com/9a95c1376e76363c1401fee7d3717173","id":"2455803089","market":"us","top_or_hot":-1,"title":"AI专题：从模型视角看端侧AI：模型技术持续演进 交互体验有望升级","media":"西南证券股份有...","content":"<div>\n<p>基础的构建：模型实现高效压缩是端侧AI的第一步。模型尺寸变小、同时具备较好性能，是端侧AI的前提。目前，在10B参数规模以下的模型中，7B尺寸占据主流，3B及以下小模型仍在探索，部分小模型性能正逐步接近更大参数模型，如谷歌Gemini-Nano模型在部分测试基准上接近Gemini-Pro、Meta Llama-3-8B模型表现可与Llama-2-70B匹敌。模型厂商为兼顾模型尺寸与性能，在算法优化...</p>\n\n<a href=\"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN202407301924429efd1226&s=b\">网页链接</a>\n\n</div>\n","source":"tencent","html":"<!DOCTYPE html>\n<html>\n<head>\n<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<meta name=\"viewport\" content=\"width=device-width,initial-scale=1.0,minimum-scale=1.0,maximum-scale=1.0,user-scalable=no\"/>\n<meta name=\"format-detection\" content=\"telephone=no,email=no,address=no\" />\n<title>AI专题：从模型视角看端侧AI：模型技术持续演进 交互体验有望升级</title>\n<style type=\"text/css\">\na,abbr,acronym,address,applet,article,aside,audio,b,big,blockquote,body,canvas,caption,center,cite,code,dd,del,details,dfn,div,dl,dt,\nem,embed,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,html,i,iframe,img,ins,kbd,label,legend,li,mark,menu,nav,\nobject,ol,output,p,pre,q,ruby,s,samp,section,small,span,strike,strong,sub,summary,sup,table,tbody,td,tfoot,th,thead,time,tr,tt,u,ul,var,video{ font:inherit;margin:0;padding:0;vertical-align:baseline;border:0 }\nbody{ font-size:16px; line-height:1.5; color:#999; background:transparent; }\n.wrapper{ overflow:hidden;word-break:break-all;padding:10px; }\nh1,h2{ font-weight:normal; line-height:1.35; margin-bottom:.6em; }\nh3,h4,h5,h6{ line-height:1.35; margin-bottom:1em; }\nh1{ font-size:24px; }\nh2{ font-size:20px; }\nh3{ font-size:18px; }\nh4{ font-size:16px; }\nh5{ font-size:14px; }\nh6{ font-size:12px; }\np,ul,ol,blockquote,dl,table{ margin:1.2em 0; }\nul,ol{ margin-left:2em; }\nul{ list-style:disc; }\nol{ list-style:decimal; }\nli,li p{ margin:10px 0;}\nimg{ max-width:100%;display:block;margin:0 auto 1em; }\nblockquote{ color:#B5B2B1; border-left:3px solid #aaa; padding:1em; }\nstrong,b{font-weight:bold;}\nem,i{font-style:italic;}\ntable{ width:100%;border-collapse:collapse;border-spacing:1px;margin:1em 0;font-size:.9em; }\nth,td{ padding:5px;text-align:left;border:1px solid #aaa; }\nth{ font-weight:bold;background:#5d5d5d; }\n.symbol-link{font-weight:bold;}\n/* header{ border-bottom:1px solid #494756; } */\n.title{ margin:0 0 8px;line-height:1.3;color:#ddd; }\n.meta {color:#5e5c6d;font-size:13px;margin:0 0 .5em; }\na{text-decoration:none; color:#2a4b87;}\n.meta .head { display: inline-block; overflow: hidden}\n.head .h-thumb { width: 30px; height: 30px; margin: 0; padding: 0; border-radius: 50%; float: left;}\n.head .h-content { margin: 0; padding: 0 0 0 9px; float: left;}\n.head .h-name {font-size: 13px; color: #eee; margin: 0;}\n.head .h-time {font-size: 11px; color: #7E829C; margin: 0;line-height: 11px;}\n.small {font-size: 12.5px; display: inline-block; transform: scale(0.9); -webkit-transform: scale(0.9); transform-origin: left; -webkit-transform-origin: left;}\n.smaller {font-size: 12.5px; display: inline-block; transform: scale(0.8); -webkit-transform: scale(0.8); transform-origin: left; -webkit-transform-origin: left;}\n.bt-text {font-size: 12px;margin: 1.5em 0 0 0}\n.bt-text p {margin: 0}\n</style>\n</head>\n<body>\n<div class=\"wrapper\">\n<header>\n<h2 class=\"title\">\nAI专题：从模型视角看端侧AI：模型技术持续演进 交互体验有望升级\n</h2>\n\n<h4 class=\"meta\">\n\n\n2024-07-30 00:00 北京时间&nbsp;&nbsp;&nbsp;<a href=http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN202407301924429efd1226&s=b><strong>西南证券股份有...</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>基础的构建：模型实现高效压缩是端侧AI的第一步。模型尺寸变小、同时具备较好性能，是端侧AI的前提。目前，在10B参数规模以下的模型中，7B尺寸占据主流，3B及以下小模型仍在探索，部分小模型性能正逐步接近更大参数模型，如谷歌Gemini-Nano模型在部分测试基准上接近Gemini-Pro、Meta Llama-3-8B模型表现可与Llama-2-70B匹敌。模型厂商为兼顾模型尺寸与性能，在算法优化...</p>\n\n<a href=\"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN202407301924429efd1226&s=b\">网页链接</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","isBrief":false,"type":0,"news_type":1,"symbol":"LU1119994496.HKD","symbol_name":"FIDELITY WORLD \"A\" (HKD) ACC","start_time":0,"source_url":"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN202407301924429efd1226&s=b","article_id":"2455803089","we_media_id":null,"thumbnails":[],"rights":{"source":"tencent","url":"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN202407301924429efd1226&s=b","rn_cache_url":null,"customStyle":"body{padding-top:10px;}#news_title{font-weight:bold;#titleStyle#;}#news_description span{font-size:12px;#descriptionStyle#;}.footer-note{#statement#}","selectors":".mod-LoadTzbdNews, body","filters":".relate-stock, .hot-list, .recom-box, .wx-sou","directOrigin":true},"url":"https://stock-news.laohu8.com/highlight/detail?id=2455803089","pubTimestamp":1722268800,"sourceInfo":{"source_id":"tencent","name":"腾讯"},"weMediaInfo":null,"summary":"模型尺寸变小、同时具备较好性能，是端侧AI的前提。未来，随着端侧模型、配套硬件、AI系统的持续发展，终端市场有望呈现更多可能。","collect":0,"end_time":0,"defaultTopTitle":"qq.com","property":[],"viewcount":null,"language":"zh","relate_stocks":{"LU1119994496.HKD":"FIDELITY WORLD \"A\" (HKD) ACC","LU0068578508.USD":"First Eagle Amundi International Cl AU-C USD","IE00B4YYXB79.USD":"PIMCO BALANCED INCOME AND GROWTH \"E\" (USD) ACC","LU0072462426.USD":"贝莱德全球配置 A2","IE00BN29S564.USD":"JANUS HENDERSON BALANCED \"A3\" (USD) INC","LU0719512351.SGD":"JPMorgan Funds - US Technology A (acc) SGD","IE00B1XK9C88.USD":"PINEBRIDGE US LARGE CAP RESEARCH ENHANCED \"A\" (USD) ACC","BK4166":"消费信贷","IE00BBT3K403.USD":"LEGG MASON CLEARBRIDGE TACTICAL DIVIDEND INCOME \"A(USD) ACC","LU0308772762.SGD":"Blackrock Global Allocation A2 SGD-H","LU0109391861.USD":"富兰克林美国机遇基金A Acc","IE00BJLML261.HKD":"HSBC GLOBAL EQUITY INDEX \"HCH\" (HKD) ACC","IE00BKDWB100.SGD":"PINEBRIDGE US LARGE CAP RESEARCH ENHANCED \"A5H\" (SGDHDG) ACC","IE00BSNM7G36.USD":"NEUBERGER BERMAN SYSTEMATIC GLOBAL SUSTAINABLE VALUE \"A\" (USD) ACC","LU0187121727.USD":"FIDELITY SUSTAINABLE US EQUITY \"A\" (USD) INC","LU1261432733.SGD":"Fidelity World A-ACC-SGD","BK4588":"碎股","LU0128525689.USD":"TEMPLETON GLOBAL BALANCED \"A\"(USD) ACC","IE00BKPKM429.USD":"NEUBERGER BERMAN GLOBAL SUSTAINABLE EQUITY \"A\" (USD) ACC","GOOGL":"谷歌A","GOOG":"谷歌","IE0005OL40V9.USD":"JANUS HENDERSON BALANCED \"A6M\" (USD) INC","LU0251142724.SGD":"Fidelity America A-SGD","LU0061474705.USD":"THREADNEEDLE (LUX) GLOBAL DYNAMIC REAL RETURN \"AU\" (USD) ACC","LU0130102774.USD":"Natixis Harris Associates US Equity RA USD","IE00BWXC8680.SGD":"PINEBRIDGE US LARGE CAP RESEARCH ENHANCED \"A5\" (SGD) ACC","LU0158827948.USD":"ALLIANZ GLOBAL SUSTAINABILITY \"A\" (USD) INC","SLM":"学贷美","SG9999001424.SGD":"United E-Commerce Fund SGD","LU0868494617.USD":"UBS (LUX) EQUITY SICAV - US TOTAL YIELD SUSTAINABLE \"P\" (USD) ACC","BK4548":"巴美列捷福持仓","IE00B1BXHZ80.USD":"Legg Mason ClearBridge - US Appreciation A Acc USD","LU0289960550.SGD":"AB FCP I - GLOBAL EQUITY BLEND PORTFOLIO 'A' (SGD) ACC","BK4170":"电脑硬件、储存设备及电脑周边","IE00BFSS8Q28.SGD":"Janus Henderson Balanced A Inc SGD-H","IE0034235303.USD":"PINEBRIDGE US RESEARCH ENHANCED CORE EQUITY \"A\" (USD) ACC","LU0238689110.USD":"贝莱德环球动力股票基金","IE0009356076.USD":"JANUS HENDERSON GLOBAL TECHNOLOGY AND INNOVATION \"A2\" (USD) ACC","LU1035775433.USD":"AB SICAV I - AMERICAN GROWTH PORTFOLIO \"AD\" (USD) INC","LU0348723411.USD":"ALLIANZ GLOBAL HI-TECH GROWTH \"A\" (USD) INC","IE00BLSP4239.USD":"Legg Mason ClearBridge - Tactical Dividend Income A Mdis USD Plus","LU0029864427.USD":"TEMPLETON GLOBAL \"A\" (USD) INC","QCOM":"高通","LU0433182093.SGD":"First Eagle Amundi International AS-C SGD","IE00B5TLWC47.USD":"BNY MELLON LONG-TERM GLOBAL EQUITY \"B\" (USD) ACC","IE00BFSS7M15.SGD":"Janus Henderson Balanced A Acc SGD-H","LU2247934214.USD":"FIDELITY FUNDS SUSTAINABLE FUTURE CONNECTIVITY \"A\" (USD) ACC","IE0004086264.USD":"BNY MELLON GLOBAL OPPORTUNITIES \"A\" (USD) ACC","IE00B3S45H60.SGD":"Neuberger Berman US Multicap Opportunities A Acc SGD-H"},"translate_title":"AI special topic: Looking at end-side AI from a model perspective: model technology continues to evolve and interactive experience is expected to be upgraded","themeId":null,"isJumpTheme":false,"ttsUrl":null,"symbols_score_info":{"SLM":1,"QCOM":0.9,"GOOG":0.9,"GOOGL":0.9},"content_text":"基础的构建：模型实现高效压缩是端侧AI的第一步。模型尺寸变小、同时具备较好性能，是端侧AI的前提。目前，在10B参数规模以下的模型中，7B尺寸占据主流，3B及以下小模型仍在探索，部分小模型性能正逐步接近更大参数模型，如谷歌Gemini-Nano模型在部分测试基准上接近Gemini-Pro、Meta Llama-3-8B模型表现可与Llama-2-70B匹敌。模型厂商为兼顾模型尺寸与性能，在算法优化上进行积极探索，在模型压缩技术、稀疏注意力机制、多头注意力变体等领域取得持续进展，帮助模型减少参数、降低存算需求，同时保持较好的性能，为端侧AI奠定小模型的基础。落地的关键：模型适配终端硬件是端侧AI的第二步。小语言模型（SLM）不完全等于端侧模型，在模型实现高效压缩后，需要进一步与手机硬件进行适配，帮助小模型装进终端。从众多小模型论文中可以发现，当前主要存在内存、功耗、算力三大硬件瓶颈。其中，苹果在其论文《LLM in aflash》中指出，70亿半精度参数的语言模型，完全加载进终端需要超过14GB的DRAM空间；Meta在其MobileLLM模型论文中指出，一个约有5000焦耳满电能量的iPhone，仅支持7B模型在10tokens/秒的AI生成速率下对话不足2小时。为解决以上问题，手机芯片厂商正加速推进AI芯片研发，在先进制程、内存容量及带宽、CPU和GPU性能、以及AI服务器上发力，手机品牌商也将配备更高性能的电池、散热元器件，提升整体终端硬件能力，更好地支持AI模型。体验的突破：模型助力人机交互是端侧AI的第三步。端侧模型通常能够支持用户完成AI初级任务，然而更丰富、更深度的交互体验需要UI模型、云端模型、以及系统级AI进行有力支撑。其中，UI模型可以提供手机UI界面理解的基础，云端模型能够帮助处理较为复杂的交互任务，系统级AI可以实现多种模型间的调用与协同。在AI时代下，模型的端侧意义不止于类似ChatGPT的聊天机器人软件，而在于赋能手机系统和应用交互的系统级AI，其带来的交互体验将成为影响用户换机的核心。从当前的海外合作阵营来看，可分为“苹果+OpenAI”和“谷歌+高通+三星”两大阵营。未来，随着端侧模型、配套硬件、AI系统的持续发展，终端市场有望呈现更多可能。相关标的：苹果(AAPL.O)、三星电子(005930.KS)、高通(QCOM.O)、谷歌(GOOGL.O)等。风险提示：端侧AI技术进展不及预期风险；行业竞争加剧风险；应用开发不及预期风险等。","kind":"news","is_publish_news":true,"is_publish_highlight":false,"is_publish_live":false,"is_publish_wemedia":null,"editions":null,"column":"","sentiment":"1","news_tag":"Product Release","news_rank":0,"symbols":[],"gpt_button":0},"commentList":[],"isCommentEnd":true,"newsSizeData":{"likeSize":0,"commentSize":0,"repostSize":0,"favoriteSize":0,"likeStatus":false,"favoriteStatus":false},"APP":{"userAgent":"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)","isDev":false,"isTTM":false,"deviceId":"web-server-community-laohu8-v3","version":"4.29.3","shortVersion":"4.29.3","platform":"web","vendor":"web","appName":"laohu8","isIOS":false,"isAndroid":false,"isTiger":false,"isTHS":false,"isWeiXin":false,"isWeiXinMini":false,"isWeiBo":false,"isQQ":false,"isBaiduSwan":false,"isBaiduBox":false,"isDingTalk":false,"isToutiao":false,"isOnePlus":false,"isHuaWei":false,"isXiaomi":false,"isXiaomiWebView":false,"isOppo":false,"isVivo":false,"isSamsung":false,"isMobile":false},"href":"/m/news/2455803089"}

AI专题：从模型视角看端侧AI：模型技术持续演进 交互体验有望升级

精彩评论

AI专题：从模型视角看端侧AI：模型技术持续演进交互体验有望升级