苹果发布DiffuCode-7B-cpGRPO编程AI模型：基于Qwen2.5-7B，可不按顺序生成代码_老虎社区_美港股上老虎

苹果发布DiffuCode-7B-cpGRPO编程AI模型：基于Qwen2.5-7B，可不按顺序生成代码

IT之家2025-07-05

IT之家 7 月 5 日消息，苹果公司悄然在 Hugging Face 上发布了一款名为 DiffuCode-7B-cpGRPO 的开源 AI 模型，该模型在生成代码方面具有创新特性，能够不按顺序生成代码，且性能媲美顶级开源编码模型。IT之家注：传统大语言模型（LLM）生成代码的方式，如同大多数人类阅读文本的方式，采用从左到右、从上到下顺序。这主要是因为这些 LLM 采用自回归（...

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。

暂无评论

热议股票

7x24

全部

{"i18n":{"language":"zh_CN"},"data":{"share":"https://www.laohu8.com/m/news/2549431108?lang=zh_CN&edition=full","thumbnail":"","is_english":false,"pubTime":"2025-07-05 09:33","share_image_url":"https://static.laohu8.com/e9f99090a1c2ed51c021029395664489","id":"2549431108","market":"us","top_or_hot":-1,"title":"苹果发布DiffuCode-7B-cpGRPO编程AI模型：基于Qwen2.5-7B，可不按顺序生成代码","media":"IT之家","content":"<div>\n<p>IT之家 7 月 5 日消息，苹果公司悄然在 Hugging Face 上发布了一款名为 DiffuCode-7B-cpGRPO 的开源 AI 模型，该模型在生成代码方面具有创新特性，能够不按顺序生成代码，且性能媲美顶级开源编码模型。IT之家注：传统大语言模型（LLM）生成代码的方式，如同大多数人类阅读文本的方式，采用从左到右、从上到下顺序。这主要是因为这些 LLM 采用自回归（...</p>\n\n<a href=\"https://tech.ifeng.com/c/8kjcpymfNdi\">网页链接</a>\n\n</div>\n","source":"fenghuang_stock","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>苹果发布DiffuCode-7B-cpGRPO编程AI模型：基于Qwen2.5-7B，可不按顺序生成代码</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\">\n苹果发布DiffuCode-7B-cpGRPO编程AI模型：基于Qwen2.5-7B，可不按顺序生成代码\n</h2>\n\n<h4 class=\"meta\">\n\n\n2025-07-05 09:33 北京时间&nbsp;&nbsp;&nbsp;<a href=https://tech.ifeng.com/c/8kjcpymfNdi><strong>IT之家</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>IT之家 7 月 5 日消息，苹果公司悄然在 Hugging Face 上发布了一款名为 DiffuCode-7B-cpGRPO 的开源 AI 模型，该模型在生成代码方面具有创新特性，能够不按顺序生成代码，且性能媲美顶级开源编码模型。IT之家注：传统大语言模型（LLM）生成代码的方式，如同大多数人类阅读文本的方式，采用从左到右、从上到下顺序。这主要是因为这些 LLM 采用自回归（...</p>\n\n<a href=\"https://tech.ifeng.com/c/8kjcpymfNdi\">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":"LU1046421795.USD","symbol_name":"富达环球科技A-ACC","start_time":0,"source_url":"https://tech.ifeng.com/c/8kjcpymfNdi","article_id":"2549431108","we_media_id":null,"thumbnails":[],"rights":null,"url":"https://stock-news.laohu8.com/highlight/detail?id=2549431108","pubTimestamp":1751679234,"columns":[],"sourceInfo":{"source_id":"fenghuang_stock","name":"凤凰网"},"weMediaInfo":null,"summary":"IT之家 7 月 5 日消息，苹果公司悄然在 Hugging Face 上发布了一款名为 DiffuCode-7B-cpGRPO 的开源 AI 模型，该模型在生成代码方面具有创新特性，能够不按顺序生成代码，且性能媲美顶级开源编码模型。苹果公司发布的模型名为 DiffuCode-7B-cpGRPO，它基于上月发表、名为《DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation》的论文。在主流编程跑分中，DiffuCode-7B-cpGRPO 保持了在生成代码时不严格依赖从左到右的生成方式情况下，相比较主流基于扩散的编程模型，测试得分提高了 4.4%。","collect":0,"end_time":0,"defaultTopTitle":"ifeng.com","property":[],"viewcount":null,"language":"zh","relate_stocks":{"LU1046421795.USD":"富达环球科技A-ACC","LU2471134796.USD":"INVESCO GLOBAL EQUITY INCOME ADVANTAGE \"A\" (USD) INC","IE00BSNM7G36.USD":"NEUBERGER BERMAN SYSTEMATIC GLOBAL SUSTAINABLE VALUE \"A\" (USD) ACC","LU0234572021.USD":"高盛美国核心股票组合Acc","LU0784385840.USD":"Blackrock Global Multi-Asset Income A2 USD","SGXZ31699556.SGD":"UGDP UNITED GLOBAL QUALITY GROWTH \"C\" (SGDHDG) ACC","MACW.SI":"APPLE 3xLongSG261006","LU0251132253.USD":"FIDELITY FUNDS  GLOBAL THEMATIC OPPORTUNITIES \"A\" (USD) ACC","LU0456855351.SGD":"JPMorgan Funds - Global Equity A (acc) SGD","LU2430703178.SGD":"WELLINGTON MULTI-ASSET HIGH INCOME \"AM4H\" (SGDHDG) INC","LU2108987350.USD":"UBS (LUX) EQUITY SICAV GLOBAL OPPORTUNITY SUSTAINABLE (USD) \"P\" (USD) ACC","LU2092627202.USD":"Blackrock ESG Multi-Asset A8 USD-H","IE0005OL40V9.USD":"JANUS HENDERSON BALANCED \"A6M\" (USD) INC","LU1642822529.SGD":"THREADNEEDLE (LUX) GLOBAL TECHNOLOGY \"A\" (SGD) ACC","LU1551013425.SGD":"Allianz Income and Growth Cl AMg2 DIS H2-SGD","LU1935043536.SGD":"MANULIFE GF GLOBAL MULTI-ASSET DIVERSIFIED INCOME \"AA\" (SGDHDG) INC A","LU2213496289.HKD":"ALLIANZ INCOME AND GROWTH \"AT\" (HKD) ACC","LU0719512351.SGD":"JPMorgan Funds - US Technology A (acc) SGD","LU0345770993.USD":"NINETY ONE GSF GLOBAL STRATEGIC EQUITY \"A\" (USD) INC","LU1803068979.SGD":"FTIF - Franklin Technology A (acc) SGD-H1","LU0345768153.USD":"NINETY ONE GSF GLOBAL STRATEGIC MANAGED \"A\" (USD) ACC","LU1066053197.SGD":"HSBC GIF GLOBAL EQUITY VOLATILITY FOCUSED \"AM3\" (SGDHDG) INC","LU2236285917.USD":"ALLIANZ GLOBAL INCOME \"AMG\" (USD) INC","SG9999014567.USD":"UOB UNITED INCOME FOCUS TRUST FUND (USD) ACC","LU1934455277.USD":"AB SICAV I LOW VOLATILITY TOTAL RETURN EQUITY PORT \"AD\" (USD) INC","LU2065170008.USD":"M&G (LUX) GLOBAL MAXIMA \"A\" (USD) INC","IE0004086264.USD":"BNY MELLON GLOBAL OPPORTUNITIES \"A\" (USD) ACC","LU0788109477.HKD":"BGF GLOBAL ALLOCATION \"A2\" (HKDHGD) ACC","SG9999014898.SGD":"United Global Quality Growth Fund Dis SGD","LU0965509283.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"AD\" (SGDHDG) INC","LU1720051108.HKD":"ALLIANZ GLOBAL ARTIFICIAL INTELLIGENCE \"AT\" (HKD) ACC","SG9999015358.SGD":"United Income Focus Trust Dis SGD-H","LU0289961442.SGD":"SUSTAINABLE GLOBAL THEMATIC PORTFOLIO \"AX\" (SGD) ACC","SG9999018865.SGD":"United Global Quality Growth Fd Cl Dist SGD-H","IE00BJJMRX11.SGD":"Janus Henderson Balanced A Acc SGD","LU0708995401.HKD":"FRANKLIN U.S. OPPORTUNITIES \"A\" (HKD) ACC","LU0942090050.USD":"UBS (LUX) EQUITY SICAV - US TOTAL YIELD SUSTAINABLE \"P\" (USD)  INC","LU2242650005.HKD":"FIDELITY FUNDS GLOBAL MULTI ASSET DYNAMIC \"A\" (HKD) ACC","LU0256863902.USD":"ALLIANZ US EQUITY \"AT\" (USD) ACC","IE00B19Z8W00.USD":"FTGF CLEARBRIDGE US LARGE CAP GROWTH \"A\" INC","LU0154236417.USD":"BGF US FLEXIBLE EQUITY \"A2\" ACC","LU1235294995.USD":"FIDELITY GLOBAL TECHNOLOGY \"A\" (USDHDG) ACC","LU2168564495.EUR":"AZ ALLOCATION - TREND \"AI\" (EUR) ACC","IE00BJTD4N35.SGD":"Neuberger Berman US Long Short Equity A1  Acc SGD-H","AAPL":"苹果","IE00B7KXQ091.USD":"Janus Henderson Balanced A Inc USD","LU2168564222.USD":"AZ ALLOCATION - TREND \"AAZ\" (USDHDG) ACC","LU0048584097.USD":"FIDELITY FUNDS  GLOBAL THEMATIC OPPORTUNITIES \"A\" (USD) INC","LU1242518857.USD":"FULLERTON LUX FUNDS - ASIA ABSOLUTE ALPHA \"I\" (USD) ACC","LU0689472784.USD":"安联收益及增长基金Cl AM AT Acc"},"translate_title":"Apple releases DiffuCode-7B-cpGRPO programming AI model: based on Qwen2.5-7B, code can be generated out of order","themeId":null,"isJumpTheme":false,"ttsUrl":null,"symbols_score_info":{"AAPL":1,"MACW.SI":0.6},"content_text":"IT之家 7 月 5 日消息，苹果公司悄然在 Hugging Face 上发布了一款名为 DiffuCode-7B-cpGRPO 的开源 AI 模型，该模型在生成代码方面具有创新特性，能够不按顺序生成代码，且性能媲美顶级开源编码模型。IT之家注：传统大语言模型（LLM）生成代码的方式，如同大多数人类阅读文本的方式，采用从左到右、从上到下顺序。这主要是因为这些 LLM 采用自回归（Autoregression）方式工作，意味着当用户向它们提问后，它们会处理整个问题，预测答案的第一个 token，然后带着这个 token 重新处理整个问题，预测第二个 token，以此类推。LLM 还有一个名为“温度”（Temperature）的设置，用于控制输出的随机性。在预测下一个 token 后，模型会为所有可能的选项分配概率。较低的温度意味着更有可能选择最可能的 token，而较高的温度则给予模型更多的自由，选择不太可能的 token。而另一种选择就是扩散（Diffusion）模型，这种模型通常用于图像模型。简而言之，模型从一个模糊、噪声的图像开始，迭代去除噪声，同时考虑到用户的需求，逐渐将其引导至更接近用户请求的图像。苹果公司发布的模型名为 DiffuCode-7B-cpGRPO，它基于上月发表、名为《DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation》的论文。论文描述了一个采用扩散优先策略的代码生成模型，但有一个特别之处：当采样温度从默认的 0.2 增加到 1.2 后，DiffuCoder 在生成 token 的顺序上变得更加灵活，从而摆脱了严格的从左到右的约束。更为有趣的是，苹果的这款模型建立在阿里的开源 Qwen2.5-7B 模型上，将这个模型按照 DiffuCoder 论文中的描述，改造成了一个基于扩散的 decoder，然后调整它以更好地遵循指示。完成这些后，他们又用超过 20000 个精心挑选的编码示例训练了它的另一个版本。在主流编程跑分中，DiffuCode-7B-cpGRPO 保持了在生成代码时不严格依赖从左到右的生成方式情况下，相比较主流基于扩散的编程模型，测试得分提高了 4.4%。","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":"productRelease","news_rank":0,"symbols":[],"gpt_button":0,"need_auth":false,"code":"91000000","status":"200"},"commentList":[],"hasMoreComments":false,"newsSizeData":{"likeSize":1,"commentSize":0,"repostSize":1,"favoriteSize":0,"likeStatus":false,"favoriteStatus":false},"isCrawlerRequest":true}