{"id":242016,"date":"2024-07-11T04:26:50","date_gmt":"2024-07-11T04:26:50","guid":{"rendered":"https:\/\/michigandigitalnews.com\/index.php\/2024\/07\/11\/anthropics-claude-3-haiku-now-fine-tunable-in-amazon-bedrock\/"},"modified":"2025-06-25T17:14:56","modified_gmt":"2025-06-25T17:14:56","slug":"anthropics-claude-3-haiku-now-fine-tunable-in-amazon-bedrock","status":"publish","type":"post","link":"https:\/\/michigandigitalnews.com\/index.php\/2024\/07\/11\/anthropics-claude-3-haiku-now-fine-tunable-in-amazon-bedrock\/","title":{"rendered":"Anthropic&#8217;s Claude 3 Haiku Now Fine-Tunable in Amazon Bedrock"},"content":{"rendered":"<p> [ad_1]<br \/>\n<\/p>\n<div>\n<figure class=\"figure mt-2\">&#13;<br \/>\n                        <a href=\"https:\/\/image.blockchain.news:443\/features\/EEF942587C092BBC865AE434AF7F9392163C996AC4BB6F3A474D06B1E81E0F4F.jpg\">&#13;<br \/>\n                            <img decoding=\"async\" class=\"rounded\" src=\"https:\/\/image.blockchain.news:443\/features\/EEF942587C092BBC865AE434AF7F9392163C996AC4BB6F3A474D06B1E81E0F4F.jpg\" alt=\"Anthropic's Claude 3 Haiku Now Fine-Tunable in Amazon Bedrock\"\/>&#13;<br \/>\n&#13;<br \/>\n                        <\/a>&#13;<br \/>\n                    <\/figure>\n<p>Anthropic has announced that customers can now fine-tune Claude 3 Haiku, the company&#8217;s fastest and most cost-effective model, within Amazon Bedrock. This new capability allows businesses to customize the model\u2019s knowledge and capabilities, making it more effective for specialized tasks, according to <a rel=\"nofollow\" href=\"https:\/\/www.anthropic.com\/news\/fine-tune-claude-3-haiku\">Anthropic<\/a>.<\/p>\n<h2>Overview of Fine-Tuning<\/h2>\n<p>Fine-tuning is a widely-used technique to enhance model performance by creating a customized version tailored to specific workflows. To fine-tune Claude 3 Haiku, users need to prepare a set of high-quality prompt-completion pairs, which are the ideal outputs for given tasks. The fine-tuning API, currently in preview, uses this data to create a custom Claude 3 Haiku model. Businesses can test and refine their custom model through the Amazon Bedrock console or API until it meets their performance goals and is ready for deployment.<\/p>\n<h2>Benefits<\/h2>\n<p>Fine-tuning Claude 3 Haiku offers several benefits:<\/p>\n<ul>\n<li><strong>Better results on specialized tasks:<\/strong> Enhance performance for domain-specific actions, such as classification and interactions with custom APIs, by encoding company and domain knowledge.<\/li>\n<li><strong>Faster speeds at lower cost:<\/strong> Reduce costs for production deployments and achieve faster results compared to other models like Sonnet or Opus.<\/li>\n<li><strong>Consistent, brand-aligned formatting:<\/strong> Generate consistently structured outputs tailored to specific requirements, ensuring compliance with regulatory and internal protocols.<\/li>\n<li><strong>Easy-to-use API:<\/strong> Enable companies of all sizes to innovate efficiently without extensive in-house AI expertise. Fine-tuning is accessible without deep technical knowledge.<\/li>\n<li><strong>Safe and secure:<\/strong> Keep proprietary training data within customers\u2019 AWS environment, preserving the Claude 3 model family\u2019s low risk of harmful outputs.<\/li>\n<\/ul>\n<p>Anthropic has demonstrated the effectiveness of fine-tuning by moderating online comments on internet forums, improving classification accuracy from 81.5% to 99.6% and reducing tokens per query by 85%.<\/p>\n<h2>Customer Spotlight<\/h2>\n<p>SK Telecom, one of South Korea&#8217;s largest telecommunications operators, has trained a custom Claude model to improve support workflows and enhance customer experiences by leveraging their industry-specific expertise. Eric Davis, Vice President of AI Tech Collaboration Group, noted a 73% increase in positive feedback for agent responses and a 37% improvement in key performance indicators for telecommunications-related tasks.<\/p>\n<p>Thomson Reuters, a global content and technology company, has also seen positive results with Claude 3 Haiku. Joel Hron, Head of AI and Labs at Thomson Reuters, highlighted the company&#8217;s aim to provide accurate, fast, and consistent user experiences by fine-tuning Claude around their industry expertise and specific requirements. Hron anticipates measurable improvements and faster speeds in AI results.<\/p>\n<h2>How to Get Started<\/h2>\n<p>Fine-tuning for Claude 3 Haiku in Amazon Bedrock is now available in preview in the US West (Oregon) AWS Region. Initially, text-based fine-tuning with context lengths up to 32K tokens is supported, with plans to introduce vision capabilities in the future. Additional details are available in the <a rel=\"nofollow\" href=\"https:\/\/aws.amazon.com\/blogs\/machine-learning\/fine-tune-anthropics-claude-3-haiku-in-amazon-bedrock-to-boost-model-accuracy-and-quality\/\">AWS launch blog<\/a> and the <a rel=\"nofollow\" href=\"https:\/\/docs.aws.amazon.com\/bedrock\/latest\/userguide\/custom-model-supported.html\">documentation<\/a>.<\/p>\n<p>To request access, contact your AWS account team or submit a support ticket in the <a rel=\"nofollow\" href=\"https:\/\/console.aws.amazon.com\/bedrock\/\">AWS Management Console<\/a>.<\/p>\n<p><span><i>Image source: Shutterstock<\/i><\/span>                    <!-- Divider --><\/p>\n<p>                    <!-- Divider --><\/p>\n<p>                    <!-- Author info START --><br \/>\n                    <!-- Author info END --><br \/>\n                    <!-- Divider -->\n                <\/div>\n<p>[ad_2]<br \/>\n<br \/><a href=\"https:\/\/blockchain.news\/news\/anthropics-claude-3-haiku-now-fine-tunable-in-amazon-bedrock\">Source link <\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>[ad_1] &#13; &#13; &#13; &#13; &#13; Anthropic has announced that customers can now fine-tune Claude 3 Haiku, the company&#8217;s fastest and most cost-effective model, within<\/p>\n","protected":false},"author":1,"featured_media":242017,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"_uf_show_specific_survey":0,"_uf_disable_surveys":false,"footnotes":""},"categories":[171],"tags":[],"_links":{"self":[{"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/posts\/242016"}],"collection":[{"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/comments?post=242016"}],"version-history":[{"count":0,"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/posts\/242016\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/media\/242017"}],"wp:attachment":[{"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/media?parent=242016"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/categories?post=242016"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/michigandigitalnews.com\/index.php\/wp-json\/wp\/v2\/tags?post=242016"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}