{"id":6216,"date":"2025-04-15T13:04:53","date_gmt":"2025-04-15T13:04:53","guid":{"rendered":"https:\/\/www.myshirtai.com\/?p=6216"},"modified":"2025-04-15T13:04:53","modified_gmt":"2025-04-15T13:04:53","slug":"openai-%e5%8f%91%e5%b8%83-gpt-4-1-%e7%b3%bb%e5%88%97%ef%bc%9a%e7%bc%96%e7%a0%81%e3%80%81%e6%8c%87%e4%bb%a4%e6%89%a7%e8%a1%8c%e4%b8%8e%e9%95%bf%e4%b8%8a%e4%b8%8b%e6%96%87%e8%83%bd%e5%8a%9b%e5%a4%a7","status":"publish","type":"post","link":"https:\/\/www.myshirtai.com\/pt\/archives\/6216","title":{"rendered":"OpenAI lan\u00e7a a s\u00e9rie GPT-4.1: melhorias dr\u00e1sticas na codifica\u00e7\u00e3o, execu\u00e7\u00e3o de instru\u00e7\u00f5es e contextos longos"},"content":{"rendered":"

I. Introdu\u00e7\u00e3o<\/h3>\n
Em 15 de abril de 2025, a OpenAI lan\u00e7ou oficialmente a nova s\u00e9rie de modelos GPT-4.1, incluindo GPT-4.1, GPT-4.1 mini e GPT-4.1 nano. Este lan\u00e7amento marca mais um grande avan\u00e7o no desempenho do modelo da OpenAI, na rela\u00e7\u00e3o custo-efic\u00e1cia e nas capacidades de aplica\u00e7\u00e3o no mundo real, especialmente nas \u00e1reas de tarefas de codifica\u00e7\u00e3o, cumprimento de instru\u00e7\u00f5es e processamento de contexto longo, proporcionando aos programadores melhores escolhas a pre\u00e7os e lat\u00eancias mais baixos.<\/div>\n

O GPT-4.1 mini est\u00e1 agora dispon\u00edvel no ShirtAI para utiliza\u00e7\u00e3o gratuita e ilimitada, a um clique de dist\u00e2ncia do s\u00edtio Web oficial:www.lsshirtai.com<\/a> <\/span><\/strong><\/p>\n

\"\"<\/div>\n
<\/div>\n

 <\/p>\n

Se pretender chamar como uma API<\/strong><\/strong> GPT-4.1<\/strong> consultar o s\u00edtio Web:https:\/\/coultra.blueshirtmap.com\/<\/a><\/span><\/strong><\/p>\n

\"\"<\/h3>\n

II. Saltar nas compet\u00eancias de codifica\u00e7\u00e3o: refor\u00e7ar toda a dimens\u00e3o, desde a gera\u00e7\u00e3o de c\u00f3digos at\u00e9 \u00e0 pr\u00e1tica de engenharia<\/h3>\n
No campo de batalha central do desenvolvimento de software, a s\u00e9rie GPT-4.1 mostra uma mudan\u00e7a qualitativa de \"gera\u00e7\u00e3o de fragmentos de c\u00f3digo\" para \"processamento de engenharia complexo\". Em resposta \u00e0s necessidades da engenharia de software do mundo real, o modelo atinge uma taxa de conclus\u00e3o de tarefas de 54,6% no teste SWE-bench Verified, que \u00e9 21% superior ao seu antecessor GPT-4o, e ultrapassa mesmo a vers\u00e3o de pr\u00e9-visualiza\u00e7\u00e3o GPT-4.5, ainda por lan\u00e7ar, em 26,6 pontos percentuais. Este avan\u00e7o n\u00e3o se reflecte apenas na precis\u00e3o da l\u00f3gica do c\u00f3digo, mas tamb\u00e9m na compreens\u00e3o profunda da base de c\u00f3digo multilingue - no teste de refer\u00eancia de diferen\u00e7as multilingues da Aider, o GPT-4.1 obteve o dobro da pontua\u00e7\u00e3o do GPT-4o, e pode seguir com precis\u00e3o o formato de diferen\u00e7as para produzir apenas as linhas modificadas, e controlar de forma est\u00e1vel o limite superior dos tokens de sa\u00edda em 32.768, o que reduz consideravelmente o n\u00famero de programadores. Pode seguir com precis\u00e3o o formato diff para produzir apenas as linhas modificadas e controlar de forma est\u00e1vel o limite de tokens de sa\u00edda a 32.768, o que reduz significativamente o custo de depura\u00e7\u00e3o dos programadores. No cen\u00e1rio de desenvolvimento front-end, a pontua\u00e7\u00e3o manual mostra que a probabilidade de a aplica\u00e7\u00e3o Web gerada ser favorecida em termos de funcionalidade e est\u00e9tica \u00e9 de 80%, e a capacidade de desenvolvimento full-stack ultrapassou pela primeira vez a maioria dos modelos de c\u00f3digo dedicados.<\/div>\n
Compara\u00e7\u00e3o dos indicadores principais:<\/div>\n
\n\n\n\n\n\n\n\n\n
modela\u00e7\u00e3o<\/th>\nSWE-bench Verificado<\/th>\nAider Multilingual Benchmarks<\/th>\nPontua\u00e7\u00e3o manual do desenvolvimento front-end<\/th>\nLimite superior do token de sa\u00edda<\/th>\nC\u00f3digo diff Precis\u00e3o<\/th>\n<\/tr>\n<\/thead>\n
GPT-4.1<\/td>\n54.6%<\/td>\n11.2<\/td>\n80%<\/td>\n32768<\/td>\n53%<\/td>\n<\/tr>\n
Pr\u00e9via do GPT-4.5<\/td>\n38.0%<\/td>\n7.4<\/td>\n52%<\/td>\n16384<\/td>\n45%<\/td>\n<\/tr>\n
o3-mini-alto<\/td>\n49.3%<\/td>\n9.8<\/td>\n65%<\/td>\n16384<\/td>\n60%<\/td>\n<\/tr>\n
o1<\/td>\n41.2%<\/td>\n6.1<\/td>\n48%<\/td>\n128000<\/td>\n62%<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

\"\"<\/h3>\n

\"\"<\/p>\n

 <\/p>\n

III. avan\u00e7o na execu\u00e7\u00e3o de comandos: precis\u00e3o e fiabilidade do processamento de tarefas complexas<\/h3>\n
Face a instru\u00e7\u00f5es complexas com m\u00faltiplos passos e restri\u00e7\u00f5es, o GPT-4.1 conseguiu passar de uma \"correspond\u00eancia difusa\" para uma \"execu\u00e7\u00e3o precisa\". No benchmark MultiChallenge da Scale, a sua pontua\u00e7\u00e3o de ader\u00eancia \u00e0s instru\u00e7\u00f5es atinge 38,3%, 10,5% superior \u00e0 do GPT-4o; e a sua pontua\u00e7\u00e3o no benchmark IFEval \u00e9 de 87,4%, ultrapassando largamente os 81,0% do seu antecessor. O modelo refor\u00e7a especialmente as tr\u00eas principais dificuldades de conformidade com o formato (por exemplo, estruturas aninhadas XML\/YAML), instru\u00e7\u00f5es negativas (rejeitando explicitamente pedidos sens\u00edveis) e tarefas ordenadas (executando fluxos de trabalho passo a passo), e a frequ\u00eancia de edi\u00e7\u00f5es inv\u00e1lidas nos cen\u00e1rios dif\u00edceis de sinaliza\u00e7\u00e3o desce para 2% de 9% no GPT-4o numa avalia\u00e7\u00e3o interna da OpenAI. Em v\u00e1rias rondas de di\u00e1logo, a sua coer\u00eancia contextual atinge 92%, seguindo com precis\u00e3o os detalhes exigidos nas instru\u00e7\u00f5es hist\u00f3ricas, proporcionando fiabilidade de n\u00edvel industrial para o servi\u00e7o inteligente ao cliente, fluxo de trabalho automatizado e outros cen\u00e1rios.<\/div>\n
Compara\u00e7\u00e3o dos indicadores principais:<\/div>\n
\n\n\n\n\n\n\n\n\n
modela\u00e7\u00e3o<\/th>\nDesafio m\u00faltiplo<\/th>\nIFEval<\/th>\nCoer\u00eancia do di\u00e1logo em v\u00e1rias rondas<\/th>\nAs diretivas negativas s\u00e3o seguidas<\/th>\nTaxa de conclus\u00e3o do mandato ordenado<\/th>\n<\/tr>\n<\/thead>\n
GPT-4.1<\/td>\n38.3%<\/td>\n87.4%<\/td>\n92%<\/td>\n98%<\/td>\n95%<\/td>\n<\/tr>\n
Pr\u00e9via do GPT-4.5<\/td>\n44.2%<\/td>\n81.0%<\/td>\n78%<\/td>\n89%<\/td>\n82%<\/td>\n<\/tr>\n
o3-mini-alto<\/td>\n40.1%<\/td>\n85.2%<\/td>\n88%<\/td>\n96%<\/td>\n91%<\/td>\n<\/tr>\n
o1<\/td>\n45.1%<\/td>\n87.1%<\/td>\n89%<\/td>\n97%<\/td>\n94%<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

\"\"<\/h3>\n

 <\/p>\n

Inova\u00e7\u00e3o de contexto longo: milh\u00f5es de janelas de fichas abrem novas possibilidades para aplica\u00e7\u00f5es profundas em v\u00e1rios cen\u00e1rios.<\/h3>\n
O GPT-4.1 vem de f\u00e1brica com uma janela de contexto de 1 milh\u00e3o de tokens, o que leva o processamento de texto longo a uma nova dimens\u00e3o - ele pode conter cerca de 8 bases de c\u00f3digo React completas ou 3.000 p\u00e1ginas de documentos legais, o que resolve completamente o problema de \"fora de contexto\" no modelo anterior. \"O problema dos modelos anteriores. Na tarefa de an\u00e1lise de v\u00eddeos longos sem gui\u00e3o Video-MME, o modelo obteve 72%, uma melhoria de 6,7% em rela\u00e7\u00e3o ao GPT-4o; os testes no conjunto de dados de c\u00f3digo aberto Graphwalks mostraram que a sua precis\u00e3o de infer\u00eancia multi-hop numa escala de milh\u00f5es de tokens atingiu 61,7%, excedendo largamente a do modelo o1 que se baseia em contextos curtos (48,7%). O OpenAI optimiza de forma s\u00edncrona a economia dos pedidos de contextos longos: uma janela de 1 milh\u00e3o de tokens \u00e9 inclu\u00edda no pre\u00e7o padr\u00e3o, o desconto da cache \u00e9 aumentado de 50% para 75% e a lat\u00eancia de resposta de 128K tokens \u00e9 reduzida para 15 segundos, o que \u00e9 30% mais r\u00e1pido do que o GPT-4.5, fornecendo uma solu\u00e7\u00e3o t\u00e9cnica fundamentada para cen\u00e1rios como a revis\u00e3o de contratos legais e a auditoria de grandes bases de c\u00f3digo.<\/div>\n
Compara\u00e7\u00e3o dos indicadores principais:<\/div>\n
\n\n\n\n\n\n\n\n\n
modela\u00e7\u00e3o<\/th>\njanela de contexto<\/th>\nV\u00eddeo-MME sem legendas<\/th>\nRacioc\u00ednio Graphwalks<\/th>\nDesconto na cache<\/th>\nAtraso de 128K para fichas<\/th>\n<\/tr>\n<\/thead>\n
GPT-4.1<\/td>\n1,000,000<\/td>\n72.0%<\/td>\n61.7%<\/td>\n75%<\/td>\n15 segundos.<\/td>\n<\/tr>\n
Pr\u00e9via do GPT-4.5<\/td>\n128,000<\/td>\n65.3%<\/td>\n42.0%<\/td>\n50%<\/td>\n22 segundos.<\/td>\n<\/tr>\n
o3-mini-alto<\/td>\n256,000<\/td>\n68.5%<\/td>\n55.2%<\/td>\n50%<\/td>\n18 segundos.<\/td>\n<\/tr>\n
o1<\/td>\n128,000<\/td>\n64.1%<\/td>\n48.7%<\/td>\n50%<\/td>\n25 segundos.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

\"\"<\/h3>\n

\"\"<\/p>\n

V. Custo e efici\u00eancia: uma atualiza\u00e7\u00e3o pragm\u00e1tica para os criadores<\/h3>\n
A estrat\u00e9gia de \"pre\u00e7os diferenciados + otimiza\u00e7\u00e3o do desempenho\" da OpenAI permite que os programadores de todas as dimens\u00f5es obtenham uma op\u00e7\u00e3o rent\u00e1vel. O modelo de entrada de gama, GPT-4.1 nano, reduz o custo de entrada para $2\/milh\u00e3o de tokens e o custo de sa\u00edda para $8\/milh\u00e3o de tokens, mantendo uma janela de milh\u00f5es de tokens, e reduz a lat\u00eancia em 50% em compara\u00e7\u00e3o com o GPT-4o, tornando-o a escolha preferida para tarefas de carga ligeira, como a categoriza\u00e7\u00e3o de texto e o preenchimento autom\u00e1tico; o GPT-4.1 mini de gama m\u00e9dia supera o GPT-4o em cen\u00e1rios de carga m\u00e9dia, como a gera\u00e7\u00e3o de c\u00f3digo e di\u00e1logos de v\u00e1rias rondas, reduzindo o custo em 60%. O modelo de gama m\u00e9dia, GPT-4.1 mini, supera o GPT-4o na gera\u00e7\u00e3o de c\u00f3digo, di\u00e1logos multi-round e outros cen\u00e1rios de carga m\u00e9dia, ao mesmo tempo que custa menos 60%. Em compara\u00e7\u00e3o, o custo de entrada da pr\u00e9-visualiza\u00e7\u00e3o do GPT-4.5 \u00e9 t\u00e3o elevado como $75\/milh\u00f5es de tokens, o que \u00e9 apenas 1\/25 da rela\u00e7\u00e3o pre\u00e7o\/desempenho do GPT-4.1, que \u00e9 a principal raz\u00e3o pela qual ser\u00e1 descontinuado at\u00e9 julho de 2025, e o novo modelo adopta uniformemente o modelo \"GPT-4\". Al\u00e9m disso, o novo modelo adopta uma pol\u00edtica uniforme de \"sem sobretaxa para contextos longos\", o que altera completamente o problema de custos do modelo anterior quando se trata de textos longos.<\/div>\n
Compara\u00e7\u00e3o dos indicadores principais:<\/div>\n
\n\n\n\n\n\n\n\n\n\n\n
modela\u00e7\u00e3o<\/th>\nCusto de entrada ($ \/ milh\u00f5es de tokens)<\/th>\nCusto de produ\u00e7\u00e3o ($\/milh\u00e3o de tokens)<\/th>\nAtraso (ficha de 128K)<\/th>\n<\/tr>\n<\/thead>\n
GPT-4.1 nano<\/td>\n0.10<\/td>\n0.40<\/td>\n5 segundos.<\/td>\n<\/tr>\n
GPT-4.1 mini<\/td>\n0.40<\/td>\n1.60<\/td>\n8 segundos.<\/td>\n<\/tr>\n
GPT-4.1<\/td>\n2.00<\/td>\n8.00<\/td>\n15 segundos.<\/td>\n<\/tr>\n
Pr\u00e9via do GPT-4.5<\/td>\n75.0<\/td>\n150.0<\/td>\n22 segundos.<\/td>\n<\/tr>\n
o3-mini-alto<\/td>\n1.10<\/td>\n4.40<\/td>\n18 segundos.<\/td>\n<\/tr>\n
o1<\/td>\n15.00<\/td>\n60.00<\/td>\n25 segundos.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n
* :: \u00cdndice custo-desempenho = (capacidade de codifica\u00e7\u00e3o + pontua\u00e7\u00e3o do comando + janela de contexto)\/(custo + lat\u00eancia), quanto mais elevado for o valor, melhor<\/span><\/div>\n
\n

 <\/p>\n

Se quiser utilizar a conta exclusiva paga oficial GPT Plus, Claude Pro, Grok Super, pode contactar a nossa equipa de profissionais (wx: abch891) se n\u00e3o souber como carregar a sua conta.<\/strong><\/p>\n\n\n\n\n\n\n\n\n
\n

Para mais produtos, consultar<\/h4>\n<\/th>\n

\n

Ver mais em<\/h4>\n<\/th>\n<\/tr>\n<\/thead>\n

ShirtAI - Intelig\u00eancia penetrante<\/a><\/td>\nO Grande Modelo do AIGC: inaugurando uma era de dupla revolu\u00e7\u00e3o na engenharia e na ci\u00eancia - Penetrating Intelligence<\/a><\/td>\n<\/tr>\n
1:1 Restaura\u00e7\u00e3o de Claude e GPT Site oficial - AI Cloud Native<\/a><\/td>\nAplica\u00e7\u00e3o de jogos em direto Leitor de visualiza\u00e7\u00e3o de desporto HD global (recomendado) - Blueshirt Technology<\/a><\/td>\n<\/tr>\n
Servi\u00e7o de tr\u00e2nsito baseado na API oficial - API GPTMeta<\/a><\/td>\nAjuda, algu\u00e9m pode dar algumas dicas sobre como fazer perguntas no GPT? - Conhecimento<\/a><\/td>\n<\/tr>\n
Loja digital de bens virtuais globais - Global SmarTone (Feng Ling Ge)<\/a><\/td>\nQu\u00e3o poderosa \u00e9 a funcionalidade Claude airtfacts que o GPT instantaneamente n\u00e3o cheira bem? -BeepBeep<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>","protected":false},"excerpt":{"rendered":"

\u4e00\u3001\u5f15\u8a00 2025 \u5e74 4 \u6708 15 \u65e5\uff0cOpenAI \u6b63\u5f0f\u63a8\u51fa\u5168\u65b0\u7684 GPT-4.1 \u7cfb\u5217\u6a21\u578b\uff0c\u5305\u542b GPT […]<\/p>","protected":false},"author":19,"featured_media":6221,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_uag_custom_page_level_css":"","site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"disabled","footer-sml-layout":"","theme-transparent-header-meta":"default","adv-header-id-meta":"","stick-header-meta":"default","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[37],"tags":[],"class_list":["post-6216","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized-zh-cn"],"yoast_head":"\nOpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347 - \u6e17\u900f\u667a\u80fd<\/title>\n<meta name=\"description\" content=\"\u63a2\u7d22 OpenAI \u5168\u65b0 GPT-4.1 \u7cfb\u5217\uff0c\u7f16\u7801\u80fd\u529b\u5927\u5e45\u8dc3\u5347\uff0c\u6307\u4ee4\u6267\u884c\u66f4\u7cbe\u51c6\u53ef\u9760\uff0c\u957f\u4e0a\u4e0b\u6587\u5904\u7406\u5b9e\u73b0\u767e\u4e07 token \u7a97\u53e3\u7a81\u7834\u3002\u4e0e GPT-4.5 \u7b49\u591a\u6a21\u578b\u5168\u65b9\u4f4d\u5bf9\u6bd4\uff0c\u6210\u672c\u4e0e\u6548\u7387\u4f18\u5316\u663e\u8457\uff0c\u4e3a\u5f00\u53d1\u8005\u5e26\u6765\u5b9e\u7528\u4e3b\u4e49\u65b0\u9009\u62e9\u3002\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.myshirtai.com\/pt\/archives\/6216\/\" \/>\n<meta property=\"og:locale\" content=\"pt_PT\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347\" \/>\n<meta property=\"og:description\" content=\"\u63a2\u7d22 OpenAI \u5168\u65b0 GPT-4.1 \u7cfb\u5217\uff0c\u7f16\u7801\u80fd\u529b\u5927\u5e45\u8dc3\u5347\uff0c\u6307\u4ee4\u6267\u884c\u66f4\u7cbe\u51c6\u53ef\u9760\uff0c\u957f\u4e0a\u4e0b\u6587\u5904\u7406\u5b9e\u73b0\u767e\u4e07 token \u7a97\u53e3\u7a81\u7834\u3002\u4e0e GPT-4.5 \u7b49\u591a\u6a21\u578b\u5168\u65b9\u4f4d\u5bf9\u6bd4\uff0c\u6210\u672c\u4e0e\u6548\u7387\u4f18\u5316\u663e\u8457\uff0c\u4e3a\u5f00\u53d1\u8005\u5e26\u6765\u5b9e\u7528\u4e3b\u4e49\u65b0\u9009\u62e9\u3002\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.myshirtai.com\/pt\/archives\/6216\/\" \/>\n<meta property=\"og:site_name\" content=\"\u6e17\u900f\u667a\u80fd\" \/>\n<meta property=\"article:published_time\" content=\"2025-04-15T13:04:53+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1015\" \/>\n\t<meta property=\"og:image:height\" content=\"691\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"\u5c0f\u9f99\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Escrito por\" \/>\n\t<meta name=\"twitter:data1\" content=\"\u5c0f\u9f99\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tempo estimado de leitura\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minutos\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216\"},\"author\":{\"name\":\"\u5c0f\u9f99\",\"@id\":\"https:\/\/www.myshirtai.com\/#\/schema\/person\/480197cd2db95877a3b4d301ffaf3777\"},\"headline\":\"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347\",\"datePublished\":\"2025-04-15T13:04:53+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216\"},\"wordCount\":145,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/www.myshirtai.com\/#organization\"},\"image\":{\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png\",\"articleSection\":[\"Uncategorized\"],\"inLanguage\":\"pt-PT\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/www.myshirtai.com\/archives\/6216#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216\",\"url\":\"https:\/\/www.myshirtai.com\/archives\/6216\",\"name\":\"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347 - \u6e17\u900f\u667a\u80fd\",\"isPartOf\":{\"@id\":\"https:\/\/www.myshirtai.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png\",\"datePublished\":\"2025-04-15T13:04:53+00:00\",\"description\":\"\u63a2\u7d22 OpenAI \u5168\u65b0 GPT-4.1 \u7cfb\u5217\uff0c\u7f16\u7801\u80fd\u529b\u5927\u5e45\u8dc3\u5347\uff0c\u6307\u4ee4\u6267\u884c\u66f4\u7cbe\u51c6\u53ef\u9760\uff0c\u957f\u4e0a\u4e0b\u6587\u5904\u7406\u5b9e\u73b0\u767e\u4e07 token \u7a97\u53e3\u7a81\u7834\u3002\u4e0e GPT-4.5 \u7b49\u591a\u6a21\u578b\u5168\u65b9\u4f4d\u5bf9\u6bd4\uff0c\u6210\u672c\u4e0e\u6548\u7387\u4f18\u5316\u663e\u8457\uff0c\u4e3a\u5f00\u53d1\u8005\u5e26\u6765\u5b9e\u7528\u4e3b\u4e49\u65b0\u9009\u62e9\u3002\",\"breadcrumb\":{\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216#breadcrumb\"},\"inLanguage\":\"pt-PT\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.myshirtai.com\/archives\/6216\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-PT\",\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage\",\"url\":\"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png\",\"contentUrl\":\"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png\",\"width\":1015,\"height\":691},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.myshirtai.com\/archives\/6216#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.myshirtai.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.myshirtai.com\/#website\",\"url\":\"https:\/\/www.myshirtai.com\/\",\"name\":\"\u6e17\u900f\u667a\u80fd\",\"description\":\"ShirtAI\",\"publisher\":{\"@id\":\"https:\/\/www.myshirtai.com\/#organization\"},\"alternateName\":\"ShirtAI\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.myshirtai.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"pt-PT\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.myshirtai.com\/#organization\",\"name\":\"ShirtAI\",\"alternateName\":\"ShirtAI\",\"url\":\"https:\/\/www.myshirtai.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-PT\",\"@id\":\"https:\/\/www.myshirtai.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2023\/11\/ShirtAI1279\u00d7675.png\",\"contentUrl\":\"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2023\/11\/ShirtAI1279\u00d7675.png\",\"width\":1200,\"height\":675,\"caption\":\"ShirtAI\"},\"image\":{\"@id\":\"https:\/\/www.myshirtai.com\/#\/schema\/logo\/image\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.myshirtai.com\/#\/schema\/person\/480197cd2db95877a3b4d301ffaf3777\",\"name\":\"\u5c0f\u9f99\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-PT\",\"@id\":\"https:\/\/www.myshirtai.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/a1915d6c04fb0c2d5c2c90c9461834429cafef21d2f6b95676462a880256f599?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/a1915d6c04fb0c2d5c2c90c9461834429cafef21d2f6b95676462a880256f599?s=96&d=mm&r=g\",\"caption\":\"\u5c0f\u9f99\"},\"url\":\"https:\/\/www.myshirtai.com\/pt\/archives\/author\/dragon\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347 - \u6e17\u900f\u667a\u80fd","description":"\u63a2\u7d22 OpenAI \u5168\u65b0 GPT-4.1 \u7cfb\u5217\uff0c\u7f16\u7801\u80fd\u529b\u5927\u5e45\u8dc3\u5347\uff0c\u6307\u4ee4\u6267\u884c\u66f4\u7cbe\u51c6\u53ef\u9760\uff0c\u957f\u4e0a\u4e0b\u6587\u5904\u7406\u5b9e\u73b0\u767e\u4e07 token \u7a97\u53e3\u7a81\u7834\u3002\u4e0e GPT-4.5 \u7b49\u591a\u6a21\u578b\u5168\u65b9\u4f4d\u5bf9\u6bd4\uff0c\u6210\u672c\u4e0e\u6548\u7387\u4f18\u5316\u663e\u8457\uff0c\u4e3a\u5f00\u53d1\u8005\u5e26\u6765\u5b9e\u7528\u4e3b\u4e49\u65b0\u9009\u62e9\u3002","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.myshirtai.com\/pt\/archives\/6216\/","og_locale":"pt_PT","og_type":"article","og_title":"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347","og_description":"\u63a2\u7d22 OpenAI \u5168\u65b0 GPT-4.1 \u7cfb\u5217\uff0c\u7f16\u7801\u80fd\u529b\u5927\u5e45\u8dc3\u5347\uff0c\u6307\u4ee4\u6267\u884c\u66f4\u7cbe\u51c6\u53ef\u9760\uff0c\u957f\u4e0a\u4e0b\u6587\u5904\u7406\u5b9e\u73b0\u767e\u4e07 token \u7a97\u53e3\u7a81\u7834\u3002\u4e0e GPT-4.5 \u7b49\u591a\u6a21\u578b\u5168\u65b9\u4f4d\u5bf9\u6bd4\uff0c\u6210\u672c\u4e0e\u6548\u7387\u4f18\u5316\u663e\u8457\uff0c\u4e3a\u5f00\u53d1\u8005\u5e26\u6765\u5b9e\u7528\u4e3b\u4e49\u65b0\u9009\u62e9\u3002","og_url":"https:\/\/www.myshirtai.com\/pt\/archives\/6216\/","og_site_name":"\u6e17\u900f\u667a\u80fd","article_published_time":"2025-04-15T13:04:53+00:00","og_image":[{"width":1015,"height":691,"url":"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png","type":"image\/png"}],"author":"\u5c0f\u9f99","twitter_card":"summary_large_image","twitter_misc":{"Escrito por":"\u5c0f\u9f99","Tempo estimado de leitura":"4 minutos"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.myshirtai.com\/archives\/6216#article","isPartOf":{"@id":"https:\/\/www.myshirtai.com\/archives\/6216"},"author":{"name":"\u5c0f\u9f99","@id":"https:\/\/www.myshirtai.com\/#\/schema\/person\/480197cd2db95877a3b4d301ffaf3777"},"headline":"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347","datePublished":"2025-04-15T13:04:53+00:00","mainEntityOfPage":{"@id":"https:\/\/www.myshirtai.com\/archives\/6216"},"wordCount":145,"commentCount":0,"publisher":{"@id":"https:\/\/www.myshirtai.com\/#organization"},"image":{"@id":"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage"},"thumbnailUrl":"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png","articleSection":["Uncategorized"],"inLanguage":"pt-PT","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.myshirtai.com\/archives\/6216#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.myshirtai.com\/archives\/6216","url":"https:\/\/www.myshirtai.com\/archives\/6216","name":"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347 - \u6e17\u900f\u667a\u80fd","isPartOf":{"@id":"https:\/\/www.myshirtai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage"},"image":{"@id":"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage"},"thumbnailUrl":"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png","datePublished":"2025-04-15T13:04:53+00:00","description":"\u63a2\u7d22 OpenAI \u5168\u65b0 GPT-4.1 \u7cfb\u5217\uff0c\u7f16\u7801\u80fd\u529b\u5927\u5e45\u8dc3\u5347\uff0c\u6307\u4ee4\u6267\u884c\u66f4\u7cbe\u51c6\u53ef\u9760\uff0c\u957f\u4e0a\u4e0b\u6587\u5904\u7406\u5b9e\u73b0\u767e\u4e07 token \u7a97\u53e3\u7a81\u7834\u3002\u4e0e GPT-4.5 \u7b49\u591a\u6a21\u578b\u5168\u65b9\u4f4d\u5bf9\u6bd4\uff0c\u6210\u672c\u4e0e\u6548\u7387\u4f18\u5316\u663e\u8457\uff0c\u4e3a\u5f00\u53d1\u8005\u5e26\u6765\u5b9e\u7528\u4e3b\u4e49\u65b0\u9009\u62e9\u3002","breadcrumb":{"@id":"https:\/\/www.myshirtai.com\/archives\/6216#breadcrumb"},"inLanguage":"pt-PT","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.myshirtai.com\/archives\/6216"]}]},{"@type":"ImageObject","inLanguage":"pt-PT","@id":"https:\/\/www.myshirtai.com\/archives\/6216#primaryimage","url":"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png","contentUrl":"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png","width":1015,"height":691},{"@type":"BreadcrumbList","@id":"https:\/\/www.myshirtai.com\/archives\/6216#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.myshirtai.com\/"},{"@type":"ListItem","position":2,"name":"OpenAI \u53d1\u5e03 GPT-4.1 \u7cfb\u5217\uff1a\u7f16\u7801\u3001\u6307\u4ee4\u6267\u884c\u4e0e\u957f\u4e0a\u4e0b\u6587\u80fd\u529b\u5927\u5e45\u63d0\u5347"}]},{"@type":"WebSite","@id":"https:\/\/www.myshirtai.com\/#website","url":"https:\/\/www.myshirtai.com\/","name":"\u6e17\u900f\u667a\u80fd","description":"ShirtAI","publisher":{"@id":"https:\/\/www.myshirtai.com\/#organization"},"alternateName":"ShirtAI","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.myshirtai.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"pt-PT"},{"@type":"Organization","@id":"https:\/\/www.myshirtai.com\/#organization","name":"ShirtAI","alternateName":"ShirtAI","url":"https:\/\/www.myshirtai.com\/","logo":{"@type":"ImageObject","inLanguage":"pt-PT","@id":"https:\/\/www.myshirtai.com\/#\/schema\/logo\/image\/","url":"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2023\/11\/ShirtAI1279\u00d7675.png","contentUrl":"https:\/\/www.myshirtai.com\/wp-content\/uploads\/2023\/11\/ShirtAI1279\u00d7675.png","width":1200,"height":675,"caption":"ShirtAI"},"image":{"@id":"https:\/\/www.myshirtai.com\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/www.myshirtai.com\/#\/schema\/person\/480197cd2db95877a3b4d301ffaf3777","name":"\u5c0f\u9f99","image":{"@type":"ImageObject","inLanguage":"pt-PT","@id":"https:\/\/www.myshirtai.com\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/a1915d6c04fb0c2d5c2c90c9461834429cafef21d2f6b95676462a880256f599?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/a1915d6c04fb0c2d5c2c90c9461834429cafef21d2f6b95676462a880256f599?s=96&d=mm&r=g","caption":"\u5c0f\u9f99"},"url":"https:\/\/www.myshirtai.com\/pt\/archives\/author\/dragon"}]}},"uagb_featured_image_src":{"full":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png",1015,691,false],"thumbnail":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657-150x102.png",150,102,true],"medium":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png",1015,691,false],"medium_large":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657-768x523.png",768,523,true],"large":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png",1015,691,false],"1536x1536":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png",1015,691,false],"2048x2048":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657.png",1015,691,false],"trp-custom-language-flag":["https:\/\/www.myshirtai.com\/wp-content\/uploads\/2025\/04\/5186bb4345f9f2132cfd2b8e3a8e9657-18x12.png",18,12,true]},"uagb_author_info":{"display_name":"\u5c0f\u9f99","author_link":"https:\/\/www.myshirtai.com\/pt\/archives\/author\/dragon"},"uagb_comment_info":0,"uagb_excerpt":"\u4e00\u3001\u5f15\u8a00 2025 \u5e74 4 \u6708 15 \u65e5\uff0cOpenAI \u6b63\u5f0f\u63a8\u51fa\u5168\u65b0\u7684 GPT-4.1 \u7cfb\u5217\u6a21\u578b\uff0c\u5305\u542b GPT…","_links":{"self":[{"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/posts\/6216","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/users\/19"}],"replies":[{"embeddable":true,"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/comments?post=6216"}],"version-history":[{"count":0,"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/posts\/6216\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/media\/6221"}],"wp:attachment":[{"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/media?parent=6216"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/categories?post=6216"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.myshirtai.com\/pt\/wp-json\/wp\/v2\/tags?post=6216"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}