<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Speculative-Decoding on knowledged.to</title><link>https://knowledged.to/tags/speculative-decoding/</link><description>Recent content in Speculative-Decoding on knowledged.to</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 24 Jun 2026 17:18:44 +0000</lastBuildDate><atom:link href="https://knowledged.to/tags/speculative-decoding/index.xml" rel="self" type="application/rss+xml"/><item><title>What is Speculative Decoding?</title><link>https://knowledged.to/ai/concepts/llm-architecture/speculative-decoding/</link><pubDate>Wed, 24 Jun 2026 17:18:08 +0000</pubDate><guid>https://knowledged.to/ai/concepts/llm-architecture/speculative-decoding/</guid><description>Explains speculative decoding, which pairs a small draft model with a large target model to accelerate LLM inference without changing outputs.</description></item><item><title>Local + Frontier Model Collaboration Patterns in Open Source Harnesses</title><link>https://knowledged.to/notes/ml/local-frontier-model-collaboration-patterns/</link><pubDate>Wed, 24 Jun 2026 17:08:01 +0000</pubDate><guid>https://knowledged.to/notes/ml/local-frontier-model-collaboration-patterns/</guid><description>New file `notes/ml/local-frontier-model-collaboration-patterns.md` added to the Notes section, alphabetically positioned after &amp;#39;LLM Thinking Token Budgets&amp;#39; and before &amp;#39;GGUF Models&amp;#39;.</description></item></channel></rss>