<?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 Pythonflow</title>
		<link>https://pythonflow.com/tags/streaming/</link>
		<description>Recent content in Streaming on Pythonflow</description>
		<generator>Hugo</generator>
		<language>en-us</language>
		
			<managingEditor>alicandonmez90@gmail.com (flowfelis)</managingEditor>
		
		
			<webMaster>alicandonmez90@gmail.com (flowfelis)</webMaster>
		
		
		
			<lastBuildDate>Tue, 23 Jun 2026 07:00:00 +0100</lastBuildDate>
		
			<atom:link href="https://pythonflow.com/tags/streaming/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>LangBot Day 7: Streaming Output — Token-by-Token Responses</title>
				<link>https://pythonflow.com/posts/langbot-day-7-streaming/</link>
				<pubDate>Tue, 23 Jun 2026 07:00:00 +0100</pubDate><author>alicandonmez90@gmail.com (flowfelis)</author>
				<guid>https://pythonflow.com/posts/langbot-day-7-streaming/</guid>
				<description>&lt;p&gt;&lt;strong&gt;Recap:&lt;/strong&gt; On Day 6, LangBot gained output parsers and now responds in two modes — chat mode returns typed &lt;code&gt;LangBotResponse&lt;/code&gt; objects with mood and confidence, and list mode returns clean &lt;code&gt;list[str]&lt;/code&gt; via &lt;code&gt;CommaSeparatedListOutputParser&lt;/code&gt;. Both modes use &lt;code&gt;.invoke()&lt;/code&gt;, which means the entire response arrives at once after the model finishes generating. Today we fix that — LangBot learns to type its replies one token at a time.&lt;/p&gt;&#xA;&lt;hr&gt;&#xA;&lt;h2 id=&#34;the-problem-waiting-for-the-whole-answer&#34;&gt;The problem: waiting for the whole answer&lt;/h2&gt;&#xA;&lt;p&gt;Every interaction so far has the same rhythm:&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
