Latest Posts

2023-06-16 | 4Bit QLoRA - 13B & 33B on 24GB VRAM or less?

In this post, we delve into training LLMs like LLaMA 13b and 33b on an RTX 3090 (or less)!

About Us

Zeus Labs is an independent research group focused on developing performant and highly capable Large Language Models. Exploration of cutting-edge and novel techniques is our speciality.

We are currently growing, and anyone interested in ML is welcome to join The Zeus Labs Discord!