What Are LLM's?

LLM is a fancy name to describe what ChatGPT is, a Large Language Model. It is "large" because it has a huge amount of parameters, we're talking billions.

ChatGPT Is Not The Only LLM In Town!

In February 2023, Meta (The Facebook company) was the first company to release a free, open source LLM, that the common mortal can run at home, on their own PC. To put things in perspective, OpenAI spent billions developing ChatGPT, and then in February, Meta comes out with a LLM that people can use for free. It's not as good as ChatGPT but the open source community is working hard at fine tuning it.

Say Hello To LLAMA!

That's the name of Meta's model. There's a 65 Billion parameter model a 33 Billion, 13 Billion and 7 Billion parameter model. The bigger the number, the more memory your computer needs to run it. You should understand parameters and models by now. Re-read this page and congratulate yourself for understanding the lingo!

Say Hello To Falcon!

In June 2023, Falcon became the newest LLM in town. Yes, another LLM, totally free and open source that you can run on your computer. Developed by the Technology Innovation Institute (TII) in Abu Dhabi, United Arab Emirates, Falcon offers better performance than LLaMA. It has three variants-1B, 7B, and 40B.

So within 6 months, the open source community has been treated to two open source LLM's!