Local LLM

What is Speculative Decoding and How Does a Draft Model Work?

September 17, 2026

I was speaking to a colleague the other day about how LLMs process tokens and predict the next word (discussing the technical details of LLMs excites me), and when I brought up speculative decoding, which

Read Full Article >

Pointing LM Studio Bionic at a Codebase With Defects I Documented in a Previous Post

September 5, 2026

Back in July I built the same login portal three times with Claude Code, once through a subagent relay, once with subagents dispatched in parallel, and once with an agent team whose members could message

Read Full Article >

Hello! My name is Terence Luk and welcome to my blog.

About me

Follow me:

Subscribe to the mailing list to receive posts updates!

Dive deep into the world of cloud computing with expert insights, tips, and tutorials tailored specifically for architects, engineers, and developers.

Local LLM

What is Speculative Decoding and How Does a Draft Model Work?

Pointing LM Studio Bionic at a Codebase With Defects I Documented in a Previous Post

Hello! My name is Terence Luk and welcome to my blog.

Follow me:

Categories

Subscribe to the mailing list to receive posts updates!

Categories

Recent Posts

Test-Driving TypeSafe AI’s Jev, the System One Model

What is Speculative Decoding and How Does a Draft Model Work?