Strk's Blog

I asked a locally-downloaded LLM to compute the cost of a car trip using natural language. No calls to any cloud service, just a model queried by an inference tool running on my own machine. The Setup The machine used for this excercise is a LemurPro laptop from around 2020, has a 11th Gen Intel(R) Core(TM) i7-1165G7 @ 2.80GHz CPU and 16GB of RAM. No GPU. The Procedure I downloaded the unsloth/gemma-4-E4B-it-qat-UD-Q4_K_XL model from Hugging Face - about 4GB in size.

This blog post was written by an AI coding agent. Specifically, by opencode, a terminal-based coding assistant, running against a remote inference provider (OpenCode Zen) serving the opencode/big-pickle model. The entire process took about two minutes. Here is how it went. I opened a terminal, typed opencode, and when the prompt appeared I pasted the following: Write a blog post about writing this blog post using this coding agent and a remote inference provider.