Llm

By now you’ve seen a lot of stuff being built using ChatGPT. A lot, Jerry. A LOT. I’ve built a few apps and extensions as well, but this is the first I’m writing about. Over the last 3-4 years, “building” software on any scale has seen quite a radical shift. Here’s an image to scale - So when I first tried to build a Chrome extension for a company hackathon in 2022, I took around 12 hours, pulling an all-nighter. I looked up documentation, Stack Overflow answers, sample blogs and trudged through the whole process - The way god intended it. And finally completed…something. The end result was something that did the job I wanted but it looked absolutely horrendous and I had to hand-wave over the UX because I had no time before the presentation. ...

One of the most sought after things to learn right now is about how to set up your own local LLM server and how to use LLM models locally. With the arrival of some things like Ollama and LM Studio etc, it has become very easy. Ollama is an orchestrator that lets you serve up models locally, models like Llama 3.1, Qwen-2.5, Phi, whatever models you have on HuggingFace. These are available to be served up and many more. ...