kekePower

6 ms·

1.
by kekePower 1y ago
Just pushed a major update: a full translation layer is now live. You can serve any HTML file and automatically translate it on the fly with a local LLM. No client JS, no cloud, no bullshit. Looking for testers, bug hunters, and feedback fr
2.
by kekePower 1y ago
I’ve been building this in my spare time over the past few days. mod_muse-ai is an Apache module written in C that lets you serve HTML pages generated by an AI model (local or remote). It reads `.ai` files, merges in a system prompt and a l
3.
Storytelling using AI – 16 models tested(aimuse.blog)
1 points by kekePower 1y ago|0 comments
4.
Running Qwen3:30B MoE on an RTX 3070 laptop with Ollama(blog.kekepower.com)
18 points by kekePower 1y ago|1 comments
5.
by kekePower 1y ago
Live and learn :-) Also to be completely and totally aware of what you request and how you request it. My dad once told me that computers are only as smart as the info you give it and this is also true for LLMs.
6.
When AI Overreach Breaks Your UI – A Frustrated Developer's Rant(blog.kekepower.com)
1 points by kekePower 1y ago|2 comments
7.
by kekePower 1y ago
It offloads to system memory, but since there are "only" 3 Billion active parameters, it works surprisingly well. I've been able to run models that are up to 29GB in size, albeit very, very slow on my system with 32GB RAM.
8.
by kekePower 1y ago
I have an RTX 3070 with 8GB VRAM and for me Qwen3:30B-A3B is fast enough. It's not lightning fast, but more than adequate if you have a _little_ patience. I've found that Qwen3 is generally really good at following instructions an