{"id":10512,"date":"2024-03-05T09:22:41","date_gmt":"2024-03-05T09:22:41","guid":{"rendered":"https:\/\/dailyai.com\/?p=10512"},"modified":"2024-03-05T09:31:11","modified_gmt":"2024-03-05T09:31:11","slug":"anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks","status":"publish","type":"post","link":"https:\/\/dailyai.com\/de\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/","title":{"rendered":"Anthropic ver\u00f6ffentlicht Claude 3, das das GPT-4 in Benchmarks schl\u00e4gt"},"content":{"rendered":"<p><strong>Das in San Francisco ans\u00e4ssige KI-Startup Anthropic hat sein neuestes LLM mit seiner Familie von Claude 3-Modellen ver\u00f6ffentlicht.<\/strong><\/p>\n<p>Claude 3 gibt es in drei Varianten: Haiku, Sonnet und Opus. F\u00fcr die weniger poetischen unter uns bedeutet das: klein, mittel und gro\u00df. Claude 3 Opus ist das fortschrittlichste Modell von Anthropic und das erste in der Branche, das behauptet, OpenAIs GPT-4 in einer Vielzahl von Benchmarks zu schlagen.<\/p>\n<p>GPT-4 war lange Zeit der Goldstandard, den KI-Unternehmen zum Vergleich ihrer LLM-Leistung herangezogen haben. Bei diesen Vergleichen wurden oft Worte wie \"ann\u00e4hernd\" oder \"fast\" verwendet, aber Anthropic kann endlich behaupten, die Leistungsf\u00e4higkeit von GPT-4 zu \u00fcbertreffen.<\/p>\n<p>Hier sind die Benchmarkwerte f\u00fcr Claude 3 im Vergleich zu GPT-4, GPT-3 und Gemini Ultra und Pro.<\/p>\n<figure id=\"attachment_10514\" aria-describedby=\"caption-attachment-10514\" style=\"width: 2200px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"size-full wp-image-10514\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark.webp\" alt=\"\" width=\"2200\" height=\"1954\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark.webp 2200w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-300x266.webp 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-1024x909.webp 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-768x682.webp 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-1536x1364.webp 1536w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-2048x1819.webp 2048w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-14x12.webp 14w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-370x329.webp 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-800x711.webp 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-20x18.webp 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-740x657.webp 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-1600x1421.webp 1600w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-benchmark-54x48.webp 54w\" sizes=\"auto, (max-width: 2200px) 100vw, 2200px\" \/><figcaption id=\"caption-attachment-10514\" class=\"wp-caption-text\">Claude 3 Benchmarkergebnisse im Vergleich zu GPT-4, GPT-3.5, Gemini Ultra und Gemini Pro. Quelle: Anthropic<\/figcaption><\/figure>\n<p>Es ist erw\u00e4hnenswert, dass die obigen GPT-4-Zahlen diejenigen sind, die OpenAI in seinem technischen Bericht vor der Ver\u00f6ffentlichung von GPT-4 vorgelegt hat. Die <a href=\"https:\/\/www-cdn.anthropic.com\/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627\/Model_Card_Claude_3.pdf\" target=\"_blank\" rel=\"noopener\">Claude 3 Modellkarte<\/a> r\u00e4umt ein, dass h\u00f6here Werte f\u00fcr GPT-4 Turbo gemeldet wurden.<\/p>\n<p>Dennoch sind die Claude-3-Opus-Figuren eine gro\u00dfe Sache. Trotz der unvermeidlichen Diskussionen dar\u00fcber, wie das Unternehmen zu diesen Zahlen gekommen ist, sagt Anthropic, dass Claude 3 Opus \"eine h\u00f6here Intelligenz als jedes andere verf\u00fcgbare Modell\" darstellt.<\/p>\n<p>Die Kosten der Claude 3 Opus Input\/Output API betragen $15 \/ $75 pro Million Token. Das ist happig im Vergleich zu GPT-4 Turbo, das $10 \/ $30 kostet. Claude 3 Sonnet ($3 \/ $15) und Claude 3 Haiku ($0,25 \/ $1,25) bieten ein wirklich gutes Preis-Leistungs-Verh\u00e4ltnis, wenn man die Leistungsdaten dieser kleineren Modelle betrachtet.<\/p>\n<p>Wenn Sie Claude 3 kostenlos ausprobieren m\u00f6chten, k\u00f6nnen Sie dies auf der Website von Anthropic tun <a href=\"https:\/\/claude.ai\/\" target=\"_blank\" rel=\"noopener\">claude.ai<\/a> Chatbot, sobald sich die Server von dem Verkehrsansturm erholt haben. Er wird von Claude 3 Sonnet angetrieben, wobei zahlende Pro-Nutzer Zugang zu Opus erhalten.<\/p>\n<p>Die Modelle von Claude 3 sind nicht multimodal, aber sie haben beeindruckende Sehf\u00e4higkeiten. Sie k\u00f6nnen zwar kein Bild f\u00fcr Sie erzeugen, aber die Benchmarks zeigen, dass Opus Fotos, Diagramme, Grafiken und technische Diagramme gut analysieren kann.<\/p>\n<figure id=\"attachment_10515\" aria-describedby=\"caption-attachment-10515\" style=\"width: 2200px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"size-full wp-image-10515\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision.webp\" alt=\"\" width=\"2200\" height=\"960\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision.webp 2200w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-300x131.webp 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-1024x447.webp 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-768x335.webp 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-1536x670.webp 1536w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-2048x894.webp 2048w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-18x8.webp 18w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-370x161.webp 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-800x349.webp 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-740x323.webp 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-20x9.webp 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-1600x698.webp 1600w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-Vision-110x48.webp 110w\" sizes=\"auto, (max-width: 2200px) 100vw, 2200px\" \/><figcaption id=\"caption-attachment-10515\" class=\"wp-caption-text\">Claude 3 Bildverarbeitungsfunktionen im Vergleich zu GPT-4V, Gemini Ultra und Gemini Pro. Quelle: Anthropic<\/figcaption><\/figure>\n<p>Anthropic sagt, dass die Claude-3-Modelle in der Lage sind, Eingaben von mehr als 1 Million Token zu akzeptieren, aber f\u00fcr die meisten Benutzer wird das Kontextfenster vorerst auf 200k Token begrenzt sein. Das ist immer noch viel mehr als der 128k-Kontext von GPT-4 Turbo.<\/p>\n<p>Ein gro\u00dfes Kontextfenster ist nur dann n\u00fctzlich, wenn es mit einer guten Erkennungsleistung gekoppelt ist. Anthropic behauptet, dass Opus eine \"nahezu perfekte Erkennungsleistung bietet, die die Genauigkeit von 99% \u00fcbertrifft\".<\/p>\n<p>W\u00e4hrend des \"Nadel im Heuhaufen\"-R\u00fcckruftests von Claude 3 Opus geschah etwas Interessantes. Als ihm eine Frage gestellt wurde, die nur beantwortet werden konnte, wenn er den eingef\u00fcgten \"Nadel\"-Satz entdeckte, zeigte er an, dass er verstanden hatte, dass er getestet wurde. Beeindruckend und ein wenig be\u00e4ngstigend.<\/p>\n<figure id=\"attachment_10521\" aria-describedby=\"caption-attachment-10521\" style=\"width: 1120px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"size-full wp-image-10521\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response.png\" alt=\"\" width=\"1120\" height=\"482\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response.png 1120w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-300x129.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-1024x441.png 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-768x331.png 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-18x8.png 18w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-370x159.png 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-800x344.png 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-740x318.png 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-20x9.png 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-3-needle-in-haystack-response-112x48.png 112w\" sizes=\"auto, (max-width: 1120px) 100vw, 1120px\" \/><figcaption id=\"caption-attachment-10521\" class=\"wp-caption-text\">Claude 3 Opus erkannte, dass es getestet wurde. Quelle: X<\/figcaption><\/figure>\n<p>Anthropic ist ein gro\u00dfer Bef\u00fcrworter dessen, was es als <a href=\"https:\/\/dailyai.com\/de\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/\">\"Konstitutionelle KI\"<\/a> die darauf abzielt, die Sicherheit und Transparenz ihrer Modelle zu verbessern. Mit <a href=\"https:\/\/dailyai.com\/de\/2023\/11\/anthropic-releases-claude-2-1-with-200k-context-window\/\">Claude 2<\/a>Dieses Streben nach Sicherheit f\u00fchrte dazu, dass sich viele weigerten, auf eigentlich harmlose Aufforderungen zu reagieren.<\/p>\n<p><a href=\"https:\/\/www.anthropic.com\/news\/claude-3-family\" target=\"_blank\" rel=\"noopener\">Claude 3<\/a> ist besser in der Lage, die Nuancen der Aufforderungen zu verstehen, um besser zu entscheiden, was gegen die anthropischen Leitplanken verst\u00f6\u00dft und was nicht. Claude 3 erreicht auch eine viel bessere Genauigkeit und weniger <a href=\"https:\/\/dailyai.com\/de\/2024\/02\/generative-ai-systems-hallucinations-and-mounting-technical-debt\/\">Halluzinationen<\/a> im Vergleich zu Claude 2.1.<\/p>\n<figure id=\"attachment_10518\" aria-describedby=\"caption-attachment-10518\" style=\"width: 1016px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"size-full wp-image-10518\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal.png\" alt=\"\" width=\"1016\" height=\"592\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal.png 1016w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-300x175.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-768x447.png 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-18x10.png 18w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-370x216.png 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-800x466.png 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-20x12.png 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-740x431.png 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-2.1-vs-Claude-3-refusal-82x48.png 82w\" sizes=\"auto, (max-width: 1016px) 100vw, 1016px\" \/><figcaption id=\"caption-attachment-10518\" class=\"wp-caption-text\">Ein Beispiel f\u00fcr eine Aufforderung, die Claude 2.1 nicht beantworten will, w\u00e4hrend Claude 3 sie als sicher erkennt.<\/figcaption><\/figure>\n<p>Einige KI-Pessimisten behaupten, dass wir auf einen KI-Winter zusteuern und dass die Leistung von LLM-Modellen ein Plateau erreicht, aber Anthropic ist da anderer Meinung. Das Unternehmen sagt, dass es nicht glaubt, dass \"die Modellintelligenz irgendwo in der N\u00e4he ihrer Grenzen ist\".<\/p>\n<p>F\u00fcr die Zukunft sind mehrere interessante Erweiterungen von Claude 3 geplant, u. a. die Hinzuf\u00fcgung fortgeschrittener Agentenf\u00e4higkeiten wie die Verwendung von Werkzeugen und die interaktive Codierung (REPL).<\/p>\n<p>Der hohe Preis k\u00f6nnte dazu f\u00fchren, dass Claude 3 Opus zun\u00e4chst eher in Nischenanwendungen in der Forschung oder im professionellen Bereich eingesetzt wird. Das Preis- und Leistungsangebot von Sonnet und Haiku wird wahrscheinlich vorerst die gr\u00f6\u00dfte Verbreitung finden.<\/p>\n<p>Werden wir eine Preissenkung von OpenAI sehen? Da OpenAI bei den Benchmarks an der Spitze steht, m\u00fcssen wir kurz vor der Ank\u00fcndigung eines GPT-5 stehen.<\/p>","protected":false},"excerpt":{"rendered":"<p>Das in San Francisco ans\u00e4ssige KI-Startup Anthropic hat sein neuestes LLM mit der Familie der Claude 3-Modelle ver\u00f6ffentlicht. Claude 3 gibt es in drei Varianten: Haiku, Sonnet und Opus. F\u00fcr die weniger poetischen unter uns bedeutet das: klein, mittel und gro\u00df. Claude 3 Opus ist das fortschrittlichste Modell von Anthropic und das erste in der Branche, das behauptet, OpenAIs GPT-4 in einer Vielzahl von Benchmarks zu schlagen. GPT-4 war lange Zeit der Goldstandard, den KI-Unternehmen zum Vergleich ihrer LLM-Leistung herangezogen haben. Bei diesen Vergleichen wurden oft Worte wie \"ann\u00e4hernd\" oder \"fast\" verwendet, aber Anthropic kann endlich behaupten, dass es die F\u00e4higkeiten von GPT-4 \u00fcbertrifft.<\/p>","protected":false},"author":6,"featured_media":10516,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[],"class_list":["post-10512","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.3 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Anthropic releases Claude 3 which beats GPT-4 in benchmarks | DailyAI<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/dailyai.com\/de\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/\" \/>\n<meta property=\"og:locale\" content=\"de_DE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Anthropic releases Claude 3 which beats GPT-4 in benchmarks | DailyAI\" \/>\n<meta property=\"og:description\" content=\"San Francisco-based AI startup Anthropic has released its latest LLM with its family of Claude 3 models. Claude 3 comes in three variations, Haiku, Sonnet, and Opus. For the less poetic among us, that translates to small, medium, and large. Claude 3 Opus is Anthropic\u2019s most advanced model and is the industry\u2019s first to claim to beat OpenAI\u2019s GPT-4 in a wide range of benchmarks. GPT-4 has been the gold standard that AI companies have long used to compare their LLM performance. Those comparisons often used words like \u201capproaching\u201d or \u201cnearly\u201d, but Anthropic can finally claim to exceed GPT-4\u2019s capability.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/de\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-03-05T09:22:41+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-03-05T09:31:11+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-Anthropic.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"562\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Eugene van der Watt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"Verfasst von\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Gesch\u00e4tzte Lesezeit\" \/>\n\t<meta name=\"twitter:data2\" content=\"5\u00a0Minuten\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"Anthropic releases Claude 3 which beats GPT-4 in benchmarks\",\"datePublished\":\"2024-03-05T09:22:41+00:00\",\"dateModified\":\"2024-03-05T09:31:11+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/\"},\"wordCount\":714,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/Claude-Anthropic.jpg\",\"articleSection\":[\"Industry\"],\"inLanguage\":\"de\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/\",\"name\":\"Anthropic releases Claude 3 which beats GPT-4 in benchmarks | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/Claude-Anthropic.jpg\",\"datePublished\":\"2024-03-05T09:22:41+00:00\",\"dateModified\":\"2024-03-05T09:31:11+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/#breadcrumb\"},\"inLanguage\":\"de\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/Claude-Anthropic.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/Claude-Anthropic.jpg\",\"width\":1000,\"height\":562},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Anthropic releases Claude 3 which beats GPT-4 in benchmarks\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"name\":\"DailyAI\",\"description\":\"Your Daily Dose of AI News\",\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/dailyai.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"de\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"width\":4501,\"height\":934,\"caption\":\"DailyAI\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/DailyAIOfficial\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/dailyaiofficial\\\/\",\"https:\\\/\\\/www.youtube.com\\\/@DailyAIOfficial\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\",\"name\":\"Eugene van der Watt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"caption\":\"Eugene van der Watt\"},\"description\":\"Eugene comes from an electronic engineering background and loves all things tech. When he takes a break from consuming AI news you'll find him at the snooker table.\",\"sameAs\":[\"www.linkedin.com\\\/in\\\/eugene-van-der-watt-16828119\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/de\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Anthropic ver\u00f6ffentlicht Claude 3, das den GPT-4 in Benchmarks schl\u00e4gt | DailyAI","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:\/\/dailyai.com\/de\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/","og_locale":"de_DE","og_type":"article","og_title":"Anthropic releases Claude 3 which beats GPT-4 in benchmarks | DailyAI","og_description":"San Francisco-based AI startup Anthropic has released its latest LLM with its family of Claude 3 models. Claude 3 comes in three variations, Haiku, Sonnet, and Opus. For the less poetic among us, that translates to small, medium, and large. Claude 3 Opus is Anthropic\u2019s most advanced model and is the industry\u2019s first to claim to beat OpenAI\u2019s GPT-4 in a wide range of benchmarks. GPT-4 has been the gold standard that AI companies have long used to compare their LLM performance. Those comparisons often used words like \u201capproaching\u201d or \u201cnearly\u201d, but Anthropic can finally claim to exceed GPT-4\u2019s capability.","og_url":"https:\/\/dailyai.com\/de\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/","og_site_name":"DailyAI","article_published_time":"2024-03-05T09:22:41+00:00","article_modified_time":"2024-03-05T09:31:11+00:00","og_image":[{"width":1000,"height":562,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-Anthropic.jpg","type":"image\/jpeg"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Verfasst von":"Eugene van der Watt","Gesch\u00e4tzte Lesezeit":"5\u00a0Minuten"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"Anthropic releases Claude 3 which beats GPT-4 in benchmarks","datePublished":"2024-03-05T09:22:41+00:00","dateModified":"2024-03-05T09:31:11+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/"},"wordCount":714,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-Anthropic.jpg","articleSection":["Industry"],"inLanguage":"de"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/","url":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/","name":"Anthropic ver\u00f6ffentlicht Claude 3, das den GPT-4 in Benchmarks schl\u00e4gt | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-Anthropic.jpg","datePublished":"2024-03-05T09:22:41+00:00","dateModified":"2024-03-05T09:31:11+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/#breadcrumb"},"inLanguage":"de","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/"]}]},{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-Anthropic.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/Claude-Anthropic.jpg","width":1000,"height":562},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/03\/anthropic-releases-claude-3-which-beats-gpt-4-in-benchmarks\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Anthropic releases Claude 3 which beats GPT-4 in benchmarks"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"Ihre t\u00e4gliche Dosis an AI-Nachrichten","publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/dailyai.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"de"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","width":4501,"height":934,"caption":"DailyAI"},"image":{"@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/DailyAIOfficial","https:\/\/www.linkedin.com\/company\/dailyaiofficial\/","https:\/\/www.youtube.com\/@DailyAIOfficial"]},{"@type":"Person","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa","name":"Eugene van der Watt","image":{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","caption":"Eugene van der Watt"},"description":"Eugene kommt aus der Elektronikbranche und liebt alles, was mit Technik zu tun hat. Wenn er eine Pause vom Konsum von KI-Nachrichten einlegt, findet man ihn am Snookertisch.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/de\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/10512","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/comments?post=10512"}],"version-history":[{"count":5,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/10512\/revisions"}],"predecessor-version":[{"id":10522,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/10512\/revisions\/10522"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media\/10516"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media?parent=10512"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/categories?post=10512"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/tags?post=10512"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}