• 0 Posts
  • 15 Comments
Joined 1 year ago
cake
Cake day: November 28th, 2024

help-circle

  • Qwen coder model from Huggingface, following the instructions there to run it in llama.cpp. Once that’s up: OpenCode and use the custom OpenAI API to connect it.

    You’ll get far better results than trying to use other local options out of the box.

    There may be better models potentially but I’ve found Qwen 2.5 etc to be pretty fantastic overall, and definitely a fine option beside Claude/ChatGPT/Gemini. I’ve tested the lot and it’s usually far more down to instruction and AGENTS.md instructions/layout than it is down to just the model.












  • Now it takes four engineers, three frameworks, and a CI/CD pipeline just to change a heading. It’s inordinately complex to simply publish a webpage.

    Huh? I mean I get that compiling a webpage that includes JS may appear more complex than uploading some unchanged HTML/CSS files, but I’d still argue you should use a build system because what you want to write and what is best delivered to browsers is usually 2 different things.

    Said build systems easily make room for JS compilation in the same way you can compile SASS to CSS and say PUG or nunjucks to HTML. You’re serving 2 separate concerns if you at all care about BOTH optimisation and devx.

    Serious old grump or out of the loop vibes in this article.