{"id":69506,"date":"2025-05-03T07:49:00","date_gmt":"2025-05-02T23:49:00","guid":{"rendered":"http:\/\/www.upgrademag.com\/web\/?p=69506"},"modified":"2025-04-30T09:51:44","modified_gmt":"2025-04-30T01:51:44","slug":"alibaba-recently-launched-qwen3","status":"publish","type":"post","link":"http:\/\/www.upgrademag.com\/web\/2025\/05\/03\/alibaba-recently-launched-qwen3\/","title":{"rendered":"Alibaba recently launched\u00a0Qwen3"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\"><strong>Alibaba has launched\u00a0<a href=\"https:\/\/qwenlm.github.io\/blog\/qwen3\/\">Qwen3<\/a>, the latest generation of its open-sourced large language model (LLM) family, setting a new benchmark for AI innovation.<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The Qwen3 series features six dense&nbsp;models and&nbsp;two Mixture-of-Experts (MoE)&nbsp;models, offering developers flexibility to build next-generation applications across mobile devices, smart glasses, autonomous vehicles, robotics, and beyond.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">All Qwen3 models \u2013 including&nbsp;dense models&nbsp;(0.6B, 1.7B, 4B, 8B, 14B, and 32B parameters) and&nbsp;MoE models&nbsp;(30B with 3B active, and 235B with 22B active) \u2013 are now open-sourced and available globally.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Hybrid Reasoning Combining Thinking and Non-thinking Modes<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Qwen3 marks Alibaba\u2019s debut of&nbsp;hybrid reasoning models, combining traditional LLM capabilities with advanced, dynamic reasoning. Qwen3 models can seamlessly switch between thinking mode for complex, multi-step tasks such as mathematics, coding, and logical deduction, and non-thinking mode for fast, general-purpose responses.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For developers accessing Qwen3 through API, the model offers granular control over&nbsp;thinking duration&nbsp;(up to 38K tokens), enabling an optimized balance between&nbsp;intelligent performance&nbsp;and compute efficiency. Notably, the Qwen3-235B-A22B MoE model significantly lowers deployment costs compared to other state-of-the-art models, reinforcing Alibaba\u2019s commitment to accessible, high-performance AI.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Breakthroughs in Multilingual Skills, Agent Capabilities, Reasoning, and Human Alignment<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Trained on a massive dataset of 36 trillion tokens \u2013 double that of its predecessor Qwen2.5 &#8212; Qwen3 delivers significant advancement on reasoning, instruction following, tool use, and multilingual tasks.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Key capabilities include:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Multilingual Mastery: Supports\u00a0119 languages and dialects, with leading performance in translation and multilingual instruction-following.<\/li>\n\n\n\n<li>Advanced Agent Integration: Natively supports the Model Context Protocol (MCP) and robust function-calling, leading open-source models in complex agent-based tasks.<\/li>\n\n\n\n<li>Superior Reasoning: Surpasses previous Qwen models (QwQ in thinking mode\u00a0and\u00a0Qwen2.5 in non-thinking mode) in mathematics, coding, and logical reasoning benchmarks.<\/li>\n\n\n\n<li>Enhanced Human Alignment: Delivers more natural creative writing, role-playing, and multi-turn dialogue experiences for more\u00a0natural, engaging conversations.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Thanks to advancements in model architecture, increase in training data, and more effective training methods, Qwen3 models achieve top-tier results across industry benchmarks such as AIME25 (mathematical reasoning),&nbsp;LiveCodeBench&nbsp;(coding proficiency),&nbsp;BFCL&nbsp;(tool and function-calling capabilities), and Arena-Hard (benchmark for instruction-tuned LLMs). Additionally, to develop the hybrid reasoning model, a four-stage training process was implemented, which includes long chain-of-thought (CoT) cold start, reasoning-based reinforcement learning (RL), thinking mode fusion, and general RL.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Open Access to Drive Innovation<\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Qwen3 models are now freely available for download on Hugging Face, Github, and ModelScope, and can be explored on chat.qwen.ai. API access will soon be available through Alibaba\u2019s AI model development platform, Model Studio. Qwen3 also powers Alibaba\u2019s flagship AI super assistant application, Quark.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Since its debut, the Qwen model family has attracted over 300 million downloads worldwide. Developers have created more than 100,000 Qwen-based derivative models on Hugging Face, making Qwen one of the world\u2019s most widely adopted open-source AI model series.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>The Qwen3 series features six dense\u00a0models and\u00a0two Mixture-of-Experts (MoE)\u00a0models, offering developers flexibility to build next-generation applications across mobile devices, smart glasses, autonomous vehicles, robotics, and beyond.<\/p>\n","protected":false},"author":6,"featured_media":63698,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[19],"tags":[4259,96,2727,1656],"class_list":["post-69506","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-headlines","tag-alibaba","tag-technology","tag-technology-adaption","tag-technology-investment"],"_links":{"self":[{"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/posts\/69506","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/comments?post=69506"}],"version-history":[{"count":1,"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/posts\/69506\/revisions"}],"predecessor-version":[{"id":69507,"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/posts\/69506\/revisions\/69507"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/media\/63698"}],"wp:attachment":[{"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/media?parent=69506"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/categories?post=69506"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.upgrademag.com\/web\/wp-json\/wp\/v2\/tags?post=69506"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}