<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Streaming on Chen Kai Blog</title><link>https://www.chenk.top/en/tags/streaming/</link><description>Recent content in Streaming on Chen Kai Blog</description><generator>Hugo</generator><language>en</language><lastBuildDate>Fri, 27 Feb 2026 09:00:00 +0000</lastBuildDate><atom:link href="https://www.chenk.top/en/tags/streaming/index.xml" rel="self" type="application/rss+xml"/><item><title>Aliyun Bailian (3): Qwen-Omni for Video, Audio, and Image Understanding</title><link>https://www.chenk.top/en/aliyun-bailian/03-qwen-omni-multimodal/</link><pubDate>Fri, 27 Feb 2026 09:00:00 +0000</pubDate><guid>https://www.chenk.top/en/aliyun-bailian/03-qwen-omni-multimodal/</guid><description>&lt;p>Of all the Bailian models, Qwen-Omni has saved me the most from product-roadmap issues. &amp;ldquo;Can you tell me what&amp;rsquo;s happening in this 2-minute promo video?&amp;rdquo; used to take 3 weeks, involving frame extraction, captioning each frame, and stitching them together. With Qwen-Omni, it&amp;rsquo;s just one HTTP request. However, the documentation lacks details on some pitfalls, such as the requirement for streaming, which has cost more than one team a half-day. Let&amp;rsquo;s avoid that for you.&lt;/p></description></item><item><title>Aliyun Bailian (2): The Qwen LLM API in Production</title><link>https://www.chenk.top/en/aliyun-bailian/02-qwen-llm-api/</link><pubDate>Thu, 26 Feb 2026 09:00:00 +0000</pubDate><guid>https://www.chenk.top/en/aliyun-bailian/02-qwen-llm-api/</guid><description>&lt;p>This article in the series covers most of the production wins. While the other models are interesting, the LLMs are what every product I&amp;rsquo;ve shipped on Bailian calls every minute of every day. The official Qwen API reference is dense and complete; this article is the readable companion that guides you through it.&lt;/p>
&lt;p>&lt;figure class="article-figure">
 &lt;img src="https://blog-pic-ck.oss-cn-beijing.aliyuncs.com/posts/en/aliyun-bailian/02-qwen-llm-api/illustration_1.png" alt="Aliyun Bailian (2): The Qwen LLM API in Production — Chapter overview" loading="lazy" decoding="async" class="content-image">
 
&lt;/figure>
&lt;/p>
&lt;hr>
&lt;h2 id="pick-the-right-qwen-variant-for-the-workload" class="heading-anchor">Pick the right Qwen variant for the workload&lt;a href="#pick-the-right-qwen-variant-for-the-workload" class="heading-link" aria-label="Permalink to this section" title="Copy link to this section">#&lt;/a>
&lt;/h2>&lt;p>The Qwen family is large. Some teams overspend by defaulting to &lt;code>qwen-max&lt;/code> everywhere; others underspend on quality by defaulting to &lt;code>qwen-turbo&lt;/code>. The right answer is &amp;ldquo;match variant to job&amp;rdquo;:&lt;/p></description></item></channel></rss>