Files
zen-blog/blog/qwen3-embedding.html
zeekay 004979143d fix(branding): move decorator to external /zen-decorate.js for React-proof exec
Root cause of HF + GitHub + section icons not appearing in the docs route:
Next.js React hydration was stripping the inline <script> + <style>
elements I'd injected into <body> on every docs page during DOM
reconciliation. The script never registered its setInterval.

Move the decorator to /zen-decorate.js loaded via <script defer src>
in <head>. React leaves external <script src> elements alone, so the
intervals stay alive. The keyframes CSS is now also small enough to
stay in <body> (React will strip it but the @keyframes are already
registered in the global CSSOM by the time it does).
2026-05-19 20:08:58 -07:00

239 lines
43 KiB
HTML
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
<!DOCTYPE html><!--i_dnJM_MIpJSOCQWNJVMq--><html lang="en" class="dark"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="stylesheet" href="/_next/static/chunks/f2332aac77592f9d.css" data-precedence="next"/><link rel="preload" as="script" fetchPriority="low" href="/_next/static/chunks/e83606e8fa9cc796.js"/><script src="/_next/static/chunks/36d6595f0156cd7e.js" async=""></script><script src="/_next/static/chunks/040e9cea20a8d9c7.js" async=""></script><script src="/_next/static/chunks/d4dffb5a0973d49c.js" async=""></script><script src="/_next/static/chunks/turbopack-967f27a9fd33556d.js" async=""></script><script src="/_next/static/chunks/59d0ad1b64f8544e.js" async=""></script><script src="/_next/static/chunks/4d80e004cf4896dd.js" async=""></script><script src="/_next/static/chunks/350ee4303b732916.js" async=""></script><script src="/_next/static/chunks/8de849ca74fc071f.js" async=""></script><script src="/_next/static/chunks/e62b91212ee7f8ff.js" async=""></script><script src="/_next/static/chunks/f19fe44237e54646.js" async=""></script><script src="/_next/static/chunks/cb0a883bafeb6805.js" async=""></script><meta name="theme-color" media="(prefers-color-scheme: dark)" content="#0A0A0A"/><meta name="theme-color" media="(prefers-color-scheme: light)" content="#fff"/><title>Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models — Zen LM Blog | Zen LM</title><meta name="description" content="GITHUB HUGGING FACE MODELSCOPE DISCORD"/><meta property="og:title" content="Zen LM - Open Foundation Models"/><meta property="og:description" content="Zen AI model family by Hanzo AI. 14 frontier models from 4B to 1T+ parameters for code, reasoning, vision, and multimodal tasks."/><meta property="og:url" content="https://zenlm.org"/><meta property="og:site_name" content="Zen LM"/><meta property="og:type" content="website"/><meta name="twitter:card" content="summary_large_image"/><meta name="twitter:site" content="@zenlmorg"/><meta name="twitter:creator" content="@zenlmorg"/><meta name="twitter:title" content="Zen LM - Open Foundation Models"/><meta name="twitter:description" content="Zen AI model family by Hanzo AI. 14 frontier models from 4B to 1T+ parameters for code, reasoning, vision, and multimodal tasks."/><script src="/_next/static/chunks/a6dad97d9634a72d.js" noModule=""></script><script defer src="/zen-decorate.js"></script></head><body class="antialiased"><div hidden=""><!--$--><!--/$--></div><script>((a,b,c,d,e,f,g,h)=>{let i=document.documentElement,j=["light","dark"];function k(b){var c;(Array.isArray(a)?a:[a]).forEach(a=>{let c="class"===a,d=c&&f?e.map(a=>f[a]||a):e;c?(i.classList.remove(...d),i.classList.add(f&&f[b]?f[b]:b)):i.setAttribute(a,b)}),c=b,h&&j.includes(c)&&(i.style.colorScheme=c)}if(d)k(d);else try{let a=localStorage.getItem(b)||c,d=g&&"system"===a?window.matchMedia("(prefers-color-scheme: dark)").matches?"dark":"light":a;k(d)}catch(a){}})("class","theme","system",null,["light","dark"],null,true,true)</script><!--$--><div data-closed="" role="presentation" hidden="" style="user-select:none;-webkit-user-select:none" class="fixed inset-0 z-50 backdrop-blur-xs bg-fd-overlay data-open:animate-fd-fade-in data-closed:animate-fd-fade-out"></div><div class="bg-fd-secondary/50 p-3 empty:hidden"></div><!--/$--><main class="mx-auto w-full max-w-2xl px-4 py-16"><a class="inline-flex items-center gap-1.5 text-sm text-fd-muted-foreground hover:text-fd-foreground mb-10 transition-colors" href="/blog">← Back to Blog</a><div class="mb-8"><time class="text-xs font-mono text-fd-muted-foreground uppercase tracking-wider">June 4, 2025</time><h1 class="text-3xl font-bold mt-2 mb-3">Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models</h1><p class="text-fd-muted-foreground text-lg mb-4">GITHUB HUGGING FACE MODELSCOPE DISCORD</p><div class="flex items-center gap-3 pt-4 border-t border-fd-border"><span class="text-sm text-fd-muted-foreground">By <!-- -->Zen LM Team</span><div class="flex gap-1.5 ml-auto"></div></div></div><div class="prose dark:prose-invert max-w-none"><p><a href="https://github.com/QwenLM/Qwen3-Embedding" rel="noreferrer noopener" target="_blank">GITHUB</a> <a href="https://huggingface.co/Qwen" rel="noreferrer noopener" target="_blank">HUGGING FACE</a> <a href="https://modelscope.cn/organization/qwen" rel="noreferrer noopener" target="_blank">MODELSCOPE</a> <a href="https://discord.gg/yPEP2vHTu4" rel="noreferrer noopener" target="_blank">DISCORD</a></p>
<p>We release <strong>Qwen3 Embedding series</strong> , a new proprietary model of the Qwen model family. These models are specifically designed for <strong>text embedding</strong> , <strong>retrieval</strong> , and <strong>reranking</strong> tasks, built on the Qwen3 foundation model. Leveraging Qwen3’s robust multilingual text understanding capabilities, the series achieves state-of-the-art performance across multiple benchmarks for text embedding and reranking tasks. We have open-sourced this series of text embedding and reranking models under the Apache 2.0 license on Hugging Face and ModelScope, and published the technical report and related code on GitHub.</p>
<p><strong>Evaluation results for reranking models</strong></p>
<div class="relative overflow-auto prose-no-margin my-6"><table><thead><tr><th>Model</th><th>Param</th><th>MTEB-R</th><th>CMTEB-R</th><th>MMTEB-R</th><th>MLDR</th><th>MTEB-Code</th><th>FollowIR</th></tr></thead><tbody><tr><td><strong>Qwen3-Embedding-0.6B</strong></td><td>0.6B</td><td>61.82</td><td>71.02</td><td>64.64</td><td>50.26</td><td>75.41</td><td>5.09</td></tr><tr><td>Jina-multilingual-reranker-v2-base</td><td>0.3B</td><td>58.22</td><td>63.37</td><td>63.73</td><td>39.66</td><td>58.98</td><td>-0.68</td></tr><tr><td>gte-multilingual-reranker-base</td><td>0.3B</td><td>59.51</td><td>74.08</td><td>59.44</td><td>66.33</td><td>54.18</td><td>-1.64</td></tr><tr><td>BGE-reranker-v2-m3</td><td>0.6B</td><td>57.03</td><td>72.16</td><td>58.36</td><td>59.51</td><td>41.38</td><td>-0.01</td></tr><tr><td><strong>Qwen3-Reranker-0.6B</strong></td><td>0.6B</td><td>65.80</td><td>71.31</td><td>66.36</td><td>67.28</td><td>73.42</td><td>5.41</td></tr><tr><td><strong>Qwen3-Reranker-4B</strong></td><td>4B</td><td><strong>69.76</strong></td><td>75.94</td><td>72.74</td><td>69.97</td><td>81.20</td><td><strong>14.84</strong></td></tr><tr><td><strong>Qwen3-Reranker-8B</strong></td><td>8B</td><td>69.02</td><td><strong>77.45</strong></td><td><strong>72.94</strong></td><td><strong>70.19</strong></td><td><strong>81.22</strong></td><td>8.05</td></tr></tbody></table></div>
<blockquote>
<p><strong>Note</strong> :</p>
<ul>
<li>We use the text retrieval subsets of MTEB(eng, v2), MTEB(cmn, v1), MTEB (Multilingual) and MTEB (Code), which are denoted as MTEB-R, CMTEB-R, MMTEB-R and MTEB-Code.</li>
<li>All scores are our runs based on the top-100 candidates retrieved by dense embedding model <a href="https://huggingface.co/Qwen/Qwen3-Embedding-0.6B" rel="noreferrer noopener" target="_blank">Qwen3-Embedding-0.6B</a>.</li>
</ul>
</blockquote>
<p><strong>Key Features</strong> :</p>
<p><strong>Exceptional Versatility</strong> : The embedding model has achieved state-of-the-art performance across a wide range of downstream application evaluations. The 8B size embedding model ranks No.1 in the MTEB multilingual leaderboard (as of June 5, 2025, score <strong>70.58</strong>). The reranking models excel in text retrieval scenarios, significantly improving search relevance.</p>
<p><strong>Comprehensive Flexibility</strong> : The Qwen3 Embedding series offers a diverse range of sizes (from 0.6B to 8B) for both embedding and reranking models, catering to various use cases that prioritize efficiency and effectiveness. Developers can seamlessly combine these two modules. Additionally, the embedding model allows for flexible vector definitions across all dimensions, and both embedding and reranking models support user-defined instructions to enhance performance for specific tasks, languages, or scenarios.</p>
<p><strong>Multilingual Capability</strong> : The Qwen3 Embedding series support over 100 languages, including various programming languages, and provides robust multilingual, cross-lingual, and code retrieval capabilities.</p>
<p><strong>Model Overview</strong> :</p>
<div class="relative overflow-auto prose-no-margin my-6"><table><thead><tr><th>Model Type</th><th>Models</th><th>Size</th><th>Layers</th><th>Sequence Length</th><th>Embedding Dimension</th><th>MRL Support</th><th>Instruction Aware</th></tr></thead><tbody><tr><td><strong>Text Embedding</strong></td><td>Qwen3-Embedding-0.6B</td><td>0.6B</td><td>28</td><td>32K</td><td>1024</td><td>Yes</td><td>Yes</td></tr><tr><td>Qwen3-Embedding-4B</td><td>4B</td><td>36</td><td>32K</td><td>2560</td><td>Yes</td><td>Yes</td><td></td></tr><tr><td>Qwen3-Embedding-8B</td><td>8B</td><td>36</td><td>32K</td><td>4096</td><td>Yes</td><td>Yes</td><td></td></tr><tr><td><strong>Text Reranking</strong></td><td>Qwen3-Reranker-0.6B</td><td>0.6B</td><td>28</td><td>32K</td><td>-</td><td>-</td><td>Yes</td></tr><tr><td>Qwen3-Reranker-4B</td><td>4B</td><td>36</td><td>32K</td><td>-</td><td>-</td><td>Yes</td><td></td></tr><tr><td>Qwen3-Reranker-8B</td><td>8B</td><td>36</td><td>32K</td><td>-</td><td>-</td><td>Yes</td><td></td></tr></tbody></table></div>
<p><em>Note: “MRL Support” indicates whether the embedding model supports custom dimensions for the final embedding. “Instruction Aware” notes whether the embedding or reranking model supports customizing the input instruction according to different tasks.</em></p>
<h2 class="flex scroll-m-28 flex-row items-center gap-2" id="model-architecture"><a data-card="" href="#model-architecture" class="peer">Model Architecture</a><svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-link size-3.5 shrink-0 text-fd-muted-foreground opacity-0 transition-opacity peer-hover:opacity-100" aria-hidden="true"><path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71"></path><path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71"></path></svg></h2>
<p>Based on the Qwen3 foundation model, our Embedding and Reranking models are designed using dual-encoder and cross-encoder architectures. Through LoRA fine-tuning, we aim to fully preserve and enhance the text understanding capabilities of the base model. The Embedding model processes a single text segment as input, extracting the semantic representation by utilizing the hidden state vector corresponding to the final <code>[EOS]</code> token. In contrast, the Reranking model takes text pairs (such as user queries and candidate documents) as input, calculating and outputting a relevance score between the pairs using a cross-encoder structure.</p>
<h2 class="flex scroll-m-28 flex-row items-center gap-2" id="model-training"><a data-card="" href="#model-training" class="peer">Model Training</a><svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-link size-3.5 shrink-0 text-fd-muted-foreground opacity-0 transition-opacity peer-hover:opacity-100" aria-hidden="true"><path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71"></path><path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71"></path></svg></h2>
<p>The training framework for the Qwen3 Embedding series follows the multi-stage training paradigm established by the GTE-Qwen series. During the training of the Embedding model, we implemented a three-stage training structure: the first stage involves contrastive pre-training with a large volume of weakly supervised data; the second stage focuses on supervised training using high-quality labeled data; and the final stage integrates multiple candidate models through a merging strategy to enhance overall performance. This staged training mechanism effectively balances the model’s generalization ability and task adaptability. For the Reranking model, based on empirical validation results, we directly employed high-quality labeled data for supervised training, significantly improving training efficiency. Notably, during the first stage of weakly supervised training for the Embedding model, we developed an innovative multi-task adaptable prompt system. By leveraging the text generation capabilities of the Qwen3 foundation model, we dynamically generated weakly supervised text pairs tailored to different task types and languages. This approach addressed the limitations of traditional methods, which often relied on community forums or open-source data for text relevance pair collection, facilitating the efficient generation of large-scale weakly supervised data.</p>
<h2 class="flex scroll-m-28 flex-row items-center gap-2" id="future-work"><a data-card="" href="#future-work" class="peer">Future work</a><svg xmlns="http://www.w3.org/2000/svg" width="24" height="24" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round" class="lucide lucide-link size-3.5 shrink-0 text-fd-muted-foreground opacity-0 transition-opacity peer-hover:opacity-100" aria-hidden="true"><path d="M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71"></path><path d="M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71"></path></svg></h2>
<p>The Qwen3 Embedding series models represent a new starting point. Through ongoing optimizations of the Qwen foundation model, we will enhance the training efficiency of text embeddings and reranking models, thereby improving deployment performance across various scenarios. Additionally, we plan to expand our multimodal representation system to establish cross-modal semantic understanding capabilities. We look forward to seeing more developers explore a wider range of scenarios based on the Qwen3 Embedding series, driving deeper applications of the model across diverse contexts.</p></div></main><!--$--><!--/$--><script src="/_next/static/chunks/e83606e8fa9cc796.js" id="_R_" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0])</script><script>self.__next_f.push([1,"1:\"$Sreact.fragment\"\n2:I[106,[\"/_next/static/chunks/59d0ad1b64f8544e.js\"],\"RootProvider\"]\n3:I[53113,[\"/_next/static/chunks/4d80e004cf4896dd.js\",\"/_next/static/chunks/350ee4303b732916.js\"],\"default\"]\n4:I[73211,[\"/_next/static/chunks/4d80e004cf4896dd.js\",\"/_next/static/chunks/350ee4303b732916.js\"],\"default\"]\n5:I[10086,[\"/_next/static/chunks/59d0ad1b64f8544e.js\",\"/_next/static/chunks/8de849ca74fc071f.js\",\"/_next/static/chunks/e62b91212ee7f8ff.js\",\"/_next/static/chunks/f19fe44237e54646.js\",\"/_next/static/chunks/cb0a883bafeb6805.js\"],\"\"]\n7:I[89923,[\"/_next/static/chunks/4d80e004cf4896dd.js\",\"/_next/static/chunks/350ee4303b732916.js\"],\"OutletBoundary\"]\n8:\"$Sreact.suspense\"\na:I[89923,[\"/_next/static/chunks/4d80e004cf4896dd.js\",\"/_next/static/chunks/350ee4303b732916.js\"],\"ViewportBoundary\"]\nc:I[89923,[\"/_next/static/chunks/4d80e004cf4896dd.js\",\"/_next/static/chunks/350ee4303b732916.js\"],\"MetadataBoundary\"]\ne:I[6998,[\"/_next/static/chunks/4d80e004cf4896dd.js\",\"/_next/static/chunks/350ee4303b732916.js\"],\"default\"]\n:HL[\"/_next/static/chunks/f2332aac77592f9d.css\",\"style\"]\n"])</script><script>self.__next_f.push([1,"0:{\"P\":null,\"b\":\"i-dnJM_MIpJSOCQWNJVMq\",\"c\":[\"\",\"blog\",\"qwen3-embedding\"],\"q\":\"\",\"i\":false,\"f\":[[[\"\",{\"children\":[\"blog\",{\"children\":[[\"slug\",\"qwen3-embedding\",\"d\"],{\"children\":[\"__PAGE__\",{}]}]}]},\"$undefined\",\"$undefined\",true],[[\"$\",\"$1\",\"c\",{\"children\":[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/_next/static/chunks/f2332aac77592f9d.css\",\"precedence\":\"next\",\"crossOrigin\":\"$undefined\",\"nonce\":\"$undefined\"}],[\"$\",\"script\",\"script-0\",{\"src\":\"/_next/static/chunks/59d0ad1b64f8544e.js\",\"async\":true,\"nonce\":\"$undefined\"}]],[\"$\",\"html\",null,{\"lang\":\"en\",\"className\":\"dark\",\"suppressHydrationWarning\":true,\"children\":[\"$\",\"body\",null,{\"className\":\"antialiased\",\"children\":[\"$\",\"$L2\",null,{\"children\":[\"$\",\"$L3\",null,{\"parallelRouterKey\":\"children\",\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L4\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"main\",null,{\"className\":\"flex min-h-screen flex-col items-center justify-center px-4 text-center\",\"children\":[[\"$\",\"div\",null,{\"className\":\"mb-8 opacity-20\",\"children\":[\"$\",\"svg\",null,{\"width\":\"120\",\"height\":\"120\",\"viewBox\":\"0 0 120 120\",\"fill\":\"none\",\"aria-hidden\":\"true\",\"children\":[\"$\",\"circle\",null,{\"cx\":\"60\",\"cy\":\"60\",\"r\":\"50\",\"stroke\":\"currentColor\",\"strokeWidth\":\"3\",\"strokeLinecap\":\"round\",\"strokeDasharray\":\"280 40\"}]}]}],[\"$\",\"p\",null,{\"className\":\"text-xs font-mono uppercase tracking-[0.3em] text-fd-muted-foreground mb-4\",\"children\":\"404\"}],[\"$\",\"h1\",null,{\"className\":\"text-3xl font-semibold mb-3\",\"children\":\"Page not found\"}],[\"$\",\"p\",null,{\"className\":\"text-fd-muted-foreground max-w-sm mb-10\",\"children\":\"This page doesn't exist, or it may have moved. Try the documentation or head home.\"}],[\"$\",\"div\",null,{\"className\":\"flex flex-wrap gap-3 justify-center\",\"children\":[[\"$\",\"$L5\",null,{\"href\":\"/\",\"className\":\"inline-flex items-center gap-2 rounded-lg bg-fd-primary text-fd-primary-foreground px-4 py-2 text-sm font-medium hover:opacity-90 transition\",\"children\":\"Go home\"}],[\"$\",\"$L5\",null,{\"href\":\"/docs\",\"className\":\"inline-flex items-center gap-2 rounded-lg border border-fd-border bg-fd-background px-4 py-2 text-sm font-medium hover:bg-fd-muted transition\",\"children\":\"Documentation\"}],[\"$\",\"$L5\",null,{\"href\":\"/docs/models\",\"className\":\"inline-flex items-center gap-2 rounded-lg border border-fd-border bg-fd-background px-4 py-2 text-sm font-medium hover:bg-fd-muted transition\",\"children\":\"Browse models\"}]]}],[\"$\",\"p\",null,{\"className\":\"mt-16 text-xs font-mono text-fd-muted-foreground opacity-50\",\"children\":\"zenlm.org\"}]]}],[]],\"forbidden\":\"$undefined\",\"unauthorized\":\"$undefined\"}]}]}]}]]}],{\"children\":[[\"$\",\"$1\",\"c\",{\"children\":[null,[\"$\",\"$L3\",null,{\"parallelRouterKey\":\"children\",\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L4\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":\"$undefined\",\"forbidden\":\"$undefined\",\"unauthorized\":\"$undefined\"}]]}],{\"children\":[[\"$\",\"$1\",\"c\",{\"children\":[null,[\"$\",\"$L3\",null,{\"parallelRouterKey\":\"children\",\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L4\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":\"$undefined\",\"forbidden\":\"$undefined\",\"unauthorized\":\"$undefined\"}]]}],{\"children\":[[\"$\",\"$1\",\"c\",{\"children\":[\"$L6\",[[\"$\",\"script\",\"script-0\",{\"src\":\"/_next/static/chunks/8de849ca74fc071f.js\",\"async\":true,\"nonce\":\"$undefined\"}],[\"$\",\"script\",\"script-1\",{\"src\":\"/_next/static/chunks/e62b91212ee7f8ff.js\",\"async\":true,\"nonce\":\"$undefined\"}],[\"$\",\"script\",\"script-2\",{\"src\":\"/_next/static/chunks/f19fe44237e54646.js\",\"async\":true,\"nonce\":\"$undefined\"}],[\"$\",\"script\",\"script-3\",{\"src\":\"/_next/static/chunks/cb0a883bafeb6805.js\",\"async\":true,\"nonce\":\"$undefined\"}]],[\"$\",\"$L7\",null,{\"children\":[\"$\",\"$8\",null,{\"name\":\"Next.MetadataOutlet\",\"children\":\"$@9\"}]}]]}],{},null,false,false]},null,false,false]},null,false,false]},null,false,false],[\"$\",\"$1\",\"h\",{\"children\":[null,[\"$\",\"$La\",null,{\"children\":\"$Lb\"}],[\"$\",\"div\",null,{\"hidden\":true,\"children\":[\"$\",\"$Lc\",null,{\"children\":[\"$\",\"$8\",null,{\"name\":\"Next.Metadata\",\"children\":\"$Ld\"}]}]}],null]}],false]],\"m\":\"$undefined\",\"G\":[\"$e\",[]],\"S\":true}\n"])</script><script>self.__next_f.push([1,"f:I[48068,[\"/_next/static/chunks/59d0ad1b64f8544e.js\",\"/_next/static/chunks/8de849ca74fc071f.js\",\"/_next/static/chunks/e62b91212ee7f8ff.js\",\"/_next/static/chunks/f19fe44237e54646.js\",\"/_next/static/chunks/cb0a883bafeb6805.js\"],\"default\"]\n"])</script><script>self.__next_f.push([1,"6:[\"$\",\"main\",null,{\"className\":\"mx-auto w-full max-w-2xl px-4 py-16\",\"children\":[[\"$\",\"$L5\",null,{\"href\":\"/blog\",\"className\":\"inline-flex items-center gap-1.5 text-sm text-fd-muted-foreground hover:text-fd-foreground mb-10 transition-colors\",\"children\":\"← Back to Blog\"}],[\"$\",\"div\",null,{\"className\":\"mb-8\",\"children\":[[\"$\",\"time\",null,{\"className\":\"text-xs font-mono text-fd-muted-foreground uppercase tracking-wider\",\"children\":\"June 4, 2025\"}],[\"$\",\"h1\",null,{\"className\":\"text-3xl font-bold mt-2 mb-3\",\"children\":\"Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models\"}],[\"$\",\"p\",null,{\"className\":\"text-fd-muted-foreground text-lg mb-4\",\"children\":\"GITHUB HUGGING FACE MODELSCOPE DISCORD\"}],[\"$\",\"div\",null,{\"className\":\"flex items-center gap-3 pt-4 border-t border-fd-border\",\"children\":[[\"$\",\"span\",null,{\"className\":\"text-sm text-fd-muted-foreground\",\"children\":[\"By \",\"Zen LM Team\"]}],[\"$\",\"div\",null,{\"className\":\"flex gap-1.5 ml-auto\",\"children\":[]}]]}]]}],[\"$\",\"div\",null,{\"className\":\"prose dark:prose-invert max-w-none\",\"children\":[[\"$\",\"p\",null,{\"children\":[[\"$\",\"$Lf\",null,{\"href\":\"https://github.com/QwenLM/Qwen3-Embedding\",\"children\":\"GITHUB\"}],\" \",[\"$\",\"$Lf\",null,{\"href\":\"https://huggingface.co/Qwen\",\"children\":\"HUGGING FACE\"}],\" \",[\"$\",\"$Lf\",null,{\"href\":\"https://modelscope.cn/organization/qwen\",\"children\":\"MODELSCOPE\"}],\" \",[\"$\",\"$Lf\",null,{\"href\":\"https://discord.gg/yPEP2vHTu4\",\"children\":\"DISCORD\"}]]}],\"\\n\",[\"$\",\"p\",null,{\"children\":[\"We release \",[\"$\",\"strong\",null,{\"children\":\"Qwen3 Embedding series\"}],\" , a new proprietary model of the Qwen model family. These models are specifically designed for \",[\"$\",\"strong\",null,{\"children\":\"text embedding\"}],\" , \",[\"$\",\"strong\",null,{\"children\":\"retrieval\"}],\" , and \",[\"$\",\"strong\",null,{\"children\":\"reranking\"}],\" tasks, built on the Qwen3 foundation model. Leveraging Qwen3’s robust multilingual text understanding capabilities, the series achieves state-of-the-art performance across multiple benchmarks for text embedding and reranking tasks. We have open-sourced this series of text embedding and reranking models under the Apache 2.0 license on Hugging Face and ModelScope, and published the technical report and related code on GitHub.\"]}],\"\\n\",[\"$\",\"p\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"Evaluation results for reranking models\"}]}],\"\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\",[\"$\",\"div\",null,{\"className\":\"relative overflow-auto prose-no-margin my-6\",\"children\":[\"$\",\"table\",null,{\"children\":[[\"$\",\"thead\",null,{\"children\":[\"$\",\"tr\",null,{\"children\":[[\"$\",\"th\",null,{\"children\":\"Model\"}],[\"$\",\"th\",null,{\"children\":\"Param\"}],[\"$\",\"th\",null,{\"children\":\"MTEB-R\"}],[\"$\",\"th\",null,{\"children\":\"CMTEB-R\"}],[\"$\",\"th\",null,{\"children\":\"MMTEB-R\"}],[\"$\",\"th\",null,{\"children\":\"MLDR\"}],[\"$\",\"th\",null,{\"children\":\"MTEB-Code\"}],[\"$\",\"th\",null,{\"children\":\"FollowIR\"}]]}]}],[\"$\",\"tbody\",null,{\"children\":[[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"Qwen3-Embedding-0.6B\"}]}],[\"$\",\"td\",null,{\"children\":\"0.6B\"}],[\"$\",\"td\",null,{\"children\":\"61.82\"}],[\"$\",\"td\",null,{\"children\":\"71.02\"}],[\"$\",\"td\",null,{\"children\":\"64.64\"}],[\"$\",\"td\",null,{\"children\":\"50.26\"}],[\"$\",\"td\",null,{\"children\":\"75.41\"}],[\"$\",\"td\",null,{\"children\":\"5.09\"}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":\"Jina-multilingual-reranker-v2-base\"}],[\"$\",\"td\",null,{\"children\":\"0.3B\"}],[\"$\",\"td\",null,{\"children\":\"58.22\"}],[\"$\",\"td\",null,{\"children\":\"63.37\"}],[\"$\",\"td\",null,{\"children\":\"63.73\"}],[\"$\",\"td\",null,{\"children\":\"39.66\"}],[\"$\",\"td\",null,{\"children\":\"58.98\"}],[\"$\",\"td\",null,{\"children\":\"-0.68\"}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":\"gte-multilingual-reranker-base\"}],[\"$\",\"td\",null,{\"children\":\"0.3B\"}],[\"$\",\"td\",null,{\"children\":\"59.51\"}],[\"$\",\"td\",null,{\"children\":\"74.08\"}],[\"$\",\"td\",null,{\"children\":\"59.44\"}],[\"$\",\"td\",null,{\"children\":\"66.33\"}],[\"$\",\"td\",null,{\"children\":\"54.18\"}],[\"$\",\"td\",null,{\"children\":\"-1.64\"}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":\"BGE-reranker-v2-m3\"}],[\"$\",\"td\",null,{\"children\":\"0.6B\"}],[\"$\",\"td\",null,{\"children\":\"57.03\"}],[\"$\",\"td\",null,{\"children\":\"72.16\"}],[\"$\",\"td\",null,{\"children\":\"58.36\"}],[\"$\",\"td\",null,{\"children\":\"59.51\"}],[\"$\",\"td\",null,{\"children\":\"41.38\"}],[\"$\",\"td\",null,{\"children\":\"-0.01\"}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"Qwen3-Reranker-0.6B\"}]}],[\"$\",\"td\",null,{\"children\":\"0.6B\"}],[\"$\",\"td\",null,{\"children\":\"65.80\"}],[\"$\",\"td\",null,{\"children\":\"71.31\"}],[\"$\",\"td\",null,{\"children\":\"66.36\"}],\"$L10\",\"$L11\",\"$L12\"]}],\"$L13\",\"$L14\"]}]]}]}],\"\\n\",\"$L15\",\"\\n\",\"$L16\",\"\\n\",\"$L17\",\"\\n\",\"$L18\",\"\\n\",\"$L19\",\"\\n\",\"$L1a\",\"\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\",\"$L1b\",\"\\n\",\"$L1c\",\"\\n\",\"$L1d\",\"\\n\",\"$L1e\",\"\\n\",\"$L1f\",\"\\n\",\"$L20\",\"\\n\",\"$L21\",\"\\n\",\"$L22\"]}]]}]\n"])</script><script>self.__next_f.push([1,"10:[\"$\",\"td\",null,{\"children\":\"67.28\"}]\n11:[\"$\",\"td\",null,{\"children\":\"73.42\"}]\n12:[\"$\",\"td\",null,{\"children\":\"5.41\"}]\n13:[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"Qwen3-Reranker-4B\"}]}],[\"$\",\"td\",null,{\"children\":\"4B\"}],[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"69.76\"}]}],[\"$\",\"td\",null,{\"children\":\"75.94\"}],[\"$\",\"td\",null,{\"children\":\"72.74\"}],[\"$\",\"td\",null,{\"children\":\"69.97\"}],[\"$\",\"td\",null,{\"children\":\"81.20\"}],[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"14.84\"}]}]]}]\n14:[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"Qwen3-Reranker-8B\"}]}],[\"$\",\"td\",null,{\"children\":\"8B\"}],[\"$\",\"td\",null,{\"children\":\"69.02\"}],[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"77.45\"}]}],[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"72.94\"}]}],[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"70.19\"}]}],[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"81.22\"}]}],[\"$\",\"td\",null,{\"children\":\"8.05\"}]]}]\n15:[\"$\",\"blockquote\",null,{\"children\":[\"\\n\",[\"$\",\"p\",null,{\"children\":[[\"$\",\"strong\",null,{\"children\":\"Note\"}],\" :\"]}],\"\\n\",[\"$\",\"ul\",null,{\"children\":[\"\\n\",[\"$\",\"li\",null,{\"children\":\"We use the text retrieval subsets of MTEB(eng, v2), MTEB(cmn, v1), MTEB (Multilingual) and MTEB (Code), which are denoted as MTEB-R, CMTEB-R, MMTEB-R and MTEB-Code.\"}],\"\\n\",[\"$\",\"li\",null,{\"children\":[\"All scores are our runs based on the top-100 candidates retrieved by dense embedding model \",[\"$\",\"$Lf\",null,{\"href\":\"https://huggingface.co/Qwen/Qwen3-Embedding-0.6B\",\"children\":\"Qwen3-Embedding-0.6B\"}],\".\"]}],\"\\n\"]}],\"\\n\"]}]\n16:[\"$\",\"p\",null,{\"children\":[[\"$\",\"strong\",null,{\"children\":\"Key Features\"}],\" :\"]}]\n17:[\"$\",\"p\",null,{\"children\":[[\"$\",\"strong\",null,{\"children\":\"Exceptional Versatility\"}],\" : The embedding model has achieved state-of-the-art performance across a wide range of downstream application evaluations. The 8B size embedding model ranks No.1 in the MTEB multilingual leaderboard (as of June 5, 2025, score \",[\"$\",\"strong\",null,{\"children\":\"70.58\"}],\"). The reranking models excel in text retrieval scenarios, significantly improving search relevance.\"]}]\n18:[\"$\",\"p\",null,{\"children\":[[\"$\",\"strong\",null,{\"children\":\"Comprehensive Flexibility\"}],\" : The Qwen3 Embedding series offers a diverse range of sizes (from 0.6B to 8B) for both embedding and reranking models, catering to various use cases that prioritize efficiency and effectiveness. Developers can seamlessly combine these two modules. Additionally, the embedding model allows for flexible vector definitions across all dimensions, and both embedding and reranking models support user-defined instructions to enhance performance for specific tasks, languages, or scenarios.\"]}]\n19:[\"$\",\"p\",null,{\"children\":[[\"$\",\"strong\",null,{\"children\":\"Multilingual Capability\"}],\" : The Qwen3 Embedding series support over 100 languages, including various programming languages, and provides robust multilingual, cross-lingual, and code retrieval capabilities.\"]}]\n1a:[\"$\",\"p\",null,{\"children\":[[\"$\",\"strong\",null,{\"children\":\"Model Overview\"}],\" :\"]}]\n"])</script><script>self.__next_f.push([1,"1b:[\"$\",\"div\",null,{\"className\":\"relative overflow-auto prose-no-margin my-6\",\"children\":[\"$\",\"table\",null,{\"children\":[[\"$\",\"thead\",null,{\"children\":[\"$\",\"tr\",null,{\"children\":[[\"$\",\"th\",null,{\"children\":\"Model Type\"}],[\"$\",\"th\",null,{\"children\":\"Models\"}],[\"$\",\"th\",null,{\"children\":\"Size\"}],[\"$\",\"th\",null,{\"children\":\"Layers\"}],[\"$\",\"th\",null,{\"children\":\"Sequence Length\"}],[\"$\",\"th\",null,{\"children\":\"Embedding Dimension\"}],[\"$\",\"th\",null,{\"children\":\"MRL Support\"}],[\"$\",\"th\",null,{\"children\":\"Instruction Aware\"}]]}]}],[\"$\",\"tbody\",null,{\"children\":[[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"Text Embedding\"}]}],[\"$\",\"td\",null,{\"children\":\"Qwen3-Embedding-0.6B\"}],[\"$\",\"td\",null,{\"children\":\"0.6B\"}],[\"$\",\"td\",null,{\"children\":\"28\"}],[\"$\",\"td\",null,{\"children\":\"32K\"}],[\"$\",\"td\",null,{\"children\":\"1024\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":\"Qwen3-Embedding-4B\"}],[\"$\",\"td\",null,{\"children\":\"4B\"}],[\"$\",\"td\",null,{\"children\":\"36\"}],[\"$\",\"td\",null,{\"children\":\"32K\"}],[\"$\",\"td\",null,{\"children\":\"2560\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}],[\"$\",\"td\",null,{}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":\"Qwen3-Embedding-8B\"}],[\"$\",\"td\",null,{\"children\":\"8B\"}],[\"$\",\"td\",null,{\"children\":\"36\"}],[\"$\",\"td\",null,{\"children\":\"32K\"}],[\"$\",\"td\",null,{\"children\":\"4096\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}],[\"$\",\"td\",null,{}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":[\"$\",\"strong\",null,{\"children\":\"Text Reranking\"}]}],[\"$\",\"td\",null,{\"children\":\"Qwen3-Reranker-0.6B\"}],[\"$\",\"td\",null,{\"children\":\"0.6B\"}],[\"$\",\"td\",null,{\"children\":\"28\"}],[\"$\",\"td\",null,{\"children\":\"32K\"}],[\"$\",\"td\",null,{\"children\":\"-\"}],[\"$\",\"td\",null,{\"children\":\"-\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":\"Qwen3-Reranker-4B\"}],[\"$\",\"td\",null,{\"children\":\"4B\"}],[\"$\",\"td\",null,{\"children\":\"36\"}],[\"$\",\"td\",null,{\"children\":\"32K\"}],[\"$\",\"td\",null,{\"children\":\"-\"}],[\"$\",\"td\",null,{\"children\":\"-\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}],[\"$\",\"td\",null,{}]]}],[\"$\",\"tr\",null,{\"children\":[[\"$\",\"td\",null,{\"children\":\"Qwen3-Reranker-8B\"}],[\"$\",\"td\",null,{\"children\":\"8B\"}],[\"$\",\"td\",null,{\"children\":\"36\"}],[\"$\",\"td\",null,{\"children\":\"32K\"}],[\"$\",\"td\",null,{\"children\":\"-\"}],[\"$\",\"td\",null,{\"children\":\"-\"}],[\"$\",\"td\",null,{\"children\":\"Yes\"}],[\"$\",\"td\",null,{}]]}]]}]]}]}]\n"])</script><script>self.__next_f.push([1,"1c:[\"$\",\"p\",null,{\"children\":[\"$\",\"em\",null,{\"children\":\"Note: “MRL Support” indicates whether the embedding model supports custom dimensions for the final embedding. “Instruction Aware” notes whether the embedding or reranking model supports customizing the input instruction according to different tasks.\"}]}]\n1d:[\"$\",\"h2\",null,{\"className\":\"flex scroll-m-28 flex-row items-center gap-2\",\"id\":\"model-architecture\",\"children\":[[\"$\",\"a\",null,{\"data-card\":\"\",\"href\":\"#model-architecture\",\"className\":\"peer\",\"children\":\"Model Architecture\"}],[\"$\",\"svg\",null,{\"ref\":\"$undefined\",\"xmlns\":\"http://www.w3.org/2000/svg\",\"width\":24,\"height\":24,\"viewBox\":\"0 0 24 24\",\"fill\":\"none\",\"stroke\":\"currentColor\",\"strokeWidth\":2,\"strokeLinecap\":\"round\",\"strokeLinejoin\":\"round\",\"className\":\"lucide lucide-link size-3.5 shrink-0 text-fd-muted-foreground opacity-0 transition-opacity peer-hover:opacity-100\",\"aria-hidden\":true,\"children\":[[\"$\",\"path\",\"1cjeqo\",{\"d\":\"M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71\"}],[\"$\",\"path\",\"19qd67\",{\"d\":\"M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71\"}],\"$undefined\"]}]]}]\n1e:[\"$\",\"p\",null,{\"children\":[\"Based on the Qwen3 foundation model, our Embedding and Reranking models are designed using dual-encoder and cross-encoder architectures. Through LoRA fine-tuning, we aim to fully preserve and enhance the text understanding capabilities of the base model. The Embedding model processes a single text segment as input, extracting the semantic representation by utilizing the hidden state vector corresponding to the final \",[\"$\",\"code\",null,{\"children\":\"[EOS]\"}],\" token. In contrast, the Reranking model takes text pairs (such as user queries and candidate documents) as input, calculating and outputting a relevance score between the pairs using a cross-encoder structure.\"]}]\n1f:[\"$\",\"h2\",null,{\"className\":\"flex scroll-m-28 flex-row items-center gap-2\",\"id\":\"model-training\",\"children\":[[\"$\",\"a\",null,{\"data-card\":\"\",\"href\":\"#model-training\",\"className\":\"peer\",\"children\":\"Model Training\"}],[\"$\",\"svg\",null,{\"ref\":\"$undefined\",\"xmlns\":\"http://www.w3.org/2000/svg\",\"width\":24,\"height\":24,\"viewBox\":\"0 0 24 24\",\"fill\":\"none\",\"stroke\":\"currentColor\",\"strokeWidth\":2,\"strokeLinecap\":\"round\",\"strokeLinejoin\":\"round\",\"className\":\"lucide lucide-link size-3.5 shrink-0 text-fd-muted-foreground opacity-0 transition-opacity peer-hover:opacity-100\",\"aria-hidden\":true,\"children\":[[\"$\",\"path\",\"1cjeqo\",{\"d\":\"M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71\"}],[\"$\",\"path\",\"19qd67\",{\"d\":\"M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71\"}],\"$undefined\"]}]]}]\n23:T561,"])</script><script>self.__next_f.push([1,"The training framework for the Qwen3 Embedding series follows the multi-stage training paradigm established by the GTE-Qwen series. During the training of the Embedding model, we implemented a three-stage training structure: the first stage involves contrastive pre-training with a large volume of weakly supervised data; the second stage focuses on supervised training using high-quality labeled data; and the final stage integrates multiple candidate models through a merging strategy to enhance overall performance. This staged training mechanism effectively balances the model’s generalization ability and task adaptability. For the Reranking model, based on empirical validation results, we directly employed high-quality labeled data for supervised training, significantly improving training efficiency. Notably, during the first stage of weakly supervised training for the Embedding model, we developed an innovative multi-task adaptable prompt system. By leveraging the text generation capabilities of the Qwen3 foundation model, we dynamically generated weakly supervised text pairs tailored to different task types and languages. This approach addressed the limitations of traditional methods, which often relied on community forums or open-source data for text relevance pair collection, facilitating the efficient generation of large-scale weakly supervised data."])</script><script>self.__next_f.push([1,"20:[\"$\",\"p\",null,{\"children\":\"$23\"}]\n21:[\"$\",\"h2\",null,{\"className\":\"flex scroll-m-28 flex-row items-center gap-2\",\"id\":\"future-work\",\"children\":[[\"$\",\"a\",null,{\"data-card\":\"\",\"href\":\"#future-work\",\"className\":\"peer\",\"children\":\"Future work\"}],[\"$\",\"svg\",null,{\"ref\":\"$undefined\",\"xmlns\":\"http://www.w3.org/2000/svg\",\"width\":24,\"height\":24,\"viewBox\":\"0 0 24 24\",\"fill\":\"none\",\"stroke\":\"currentColor\",\"strokeWidth\":2,\"strokeLinecap\":\"round\",\"strokeLinejoin\":\"round\",\"className\":\"lucide lucide-link size-3.5 shrink-0 text-fd-muted-foreground opacity-0 transition-opacity peer-hover:opacity-100\",\"aria-hidden\":true,\"children\":[[\"$\",\"path\",\"1cjeqo\",{\"d\":\"M10 13a5 5 0 0 0 7.54.54l3-3a5 5 0 0 0-7.07-7.07l-1.72 1.71\"}],[\"$\",\"path\",\"19qd67\",{\"d\":\"M14 11a5 5 0 0 0-7.54-.54l-3 3a5 5 0 0 0 7.07 7.07l1.71-1.71\"}],\"$undefined\"]}]]}]\n22:[\"$\",\"p\",null,{\"children\":\"The Qwen3 Embedding series models represent a new starting point. Through ongoing optimizations of the Qwen foundation model, we will enhance the training efficiency of text embeddings and reranking models, thereby improving deployment performance across various scenarios. Additionally, we plan to expand our multimodal representation system to establish cross-modal semantic understanding capabilities. We look forward to seeing more developers explore a wider range of scenarios based on the Qwen3 Embedding series, driving deeper applications of the model across diverse contexts.\"}]\n"])</script><script>self.__next_f.push([1,"b:[[\"$\",\"meta\",\"0\",{\"charSet\":\"utf-8\"}],[\"$\",\"meta\",\"1\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"2\",{\"name\":\"theme-color\",\"media\":\"(prefers-color-scheme: dark)\",\"content\":\"#0A0A0A\"}],[\"$\",\"meta\",\"3\",{\"name\":\"theme-color\",\"media\":\"(prefers-color-scheme: light)\",\"content\":\"#fff\"}]]\n"])</script><script>self.__next_f.push([1,"9:null\nd:[[\"$\",\"title\",\"0\",{\"children\":\"Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models — Zen LM Blog | Zen LM\"}],[\"$\",\"meta\",\"1\",{\"name\":\"description\",\"content\":\"GITHUB HUGGING FACE MODELSCOPE DISCORD\"}],[\"$\",\"meta\",\"2\",{\"property\":\"og:title\",\"content\":\"Zen LM - Open Foundation Models\"}],[\"$\",\"meta\",\"3\",{\"property\":\"og:description\",\"content\":\"Zen AI model family by Hanzo AI. 14 frontier models from 4B to 1T+ parameters for code, reasoning, vision, and multimodal tasks.\"}],[\"$\",\"meta\",\"4\",{\"property\":\"og:url\",\"content\":\"https://zenlm.org\"}],[\"$\",\"meta\",\"5\",{\"property\":\"og:site_name\",\"content\":\"Zen LM\"}],[\"$\",\"meta\",\"6\",{\"property\":\"og:type\",\"content\":\"website\"}],[\"$\",\"meta\",\"7\",{\"name\":\"twitter:card\",\"content\":\"summary_large_image\"}],[\"$\",\"meta\",\"8\",{\"name\":\"twitter:site\",\"content\":\"@zenlmorg\"}],[\"$\",\"meta\",\"9\",{\"name\":\"twitter:creator\",\"content\":\"@zenlmorg\"}],[\"$\",\"meta\",\"10\",{\"name\":\"twitter:title\",\"content\":\"Zen LM - Open Foundation Models\"}],[\"$\",\"meta\",\"11\",{\"name\":\"twitter:description\",\"content\":\"Zen AI model family by Hanzo AI. 14 frontier models from 4B to 1T+ parameters for code, reasoning, vision, and multimodal tasks.\"}]]\n"])</script><script>
(function(){
// After React renders the lotus emoji from RSC payload, walk the DOM
// and replace it with the enso SVG. Polls so subsequent React re-renders
// also get cleaned up.
function swap(){
// 1. Replace lotus + ' Zen LM' text inside <a> brand links
document.querySelectorAll('a').forEach(function(a){
if (a.dataset.zenLotusSwapped) return;
var t = (a.textContent || '');
if (!/.*Zen LM|Zen LM.*/.test(t)) return;
// Find the lotus text node specifically
var walker = document.createTreeWalker(a, NodeFilter.SHOW_TEXT, null);
var node;
while ((node = walker.nextNode())) {
if (node.nodeValue && node.nodeValue.indexOf('') !== -1) {
node.nodeValue = node.nodeValue.replace(/\s?/g, '');
}
}
// Inject enso SVG at start
if (!a.querySelector('svg[aria-label="Zen LM"]')) {
var ns = 'http://www.w3.org/2000/svg';
var svg = document.createElementNS(ns, 'svg');
svg.setAttribute('viewBox', '0 0 512 512');
svg.setAttribute('fill', 'none');
svg.setAttribute('width', '18');
svg.setAttribute('height', '18');
svg.setAttribute('aria-label', 'Zen LM');
svg.style.cssText = 'display:inline-block;vertical-align:-2px;margin-right:6px;flex-shrink:0';
var p = document.createElementNS(ns, 'path');
p.setAttribute('d', 'M 256 84 A 172 172 0 1 0 256 428 A 172 172 0 1 0 256 84 Z');
p.setAttribute('stroke', 'currentColor');
p.setAttribute('stroke-width', '26');
p.setAttribute('stroke-linecap', 'round');
p.setAttribute('stroke-linejoin', 'round');
svg.appendChild(p);
a.insertBefore(svg, a.firstChild);
}
a.dataset.zenLotusSwapped = '1';
});
}
function start(){ swap(); setInterval(swap, 600); }
if (document.readyState !== 'loading') start();
else document.addEventListener('DOMContentLoaded', start);
})();
</script>
<style id="zenlm-brand-anim">
@keyframes zenlm-brand-draw { to { stroke-dashoffset: 0; } }
svg.zenlm-brand-svg .zenlm-brand-path {
stroke-dasharray: 1100;
stroke-dashoffset: 1100;
animation: zenlm-brand-draw 1.4s cubic-bezier(.45,0,.2,1) 0.1s 1 forwards;
}
.zen-link-icon { display: inline-block; vertical-align: middle; flex-shrink: 0; }
</style></body></html>