How to run large language models from Hugging Face without needing your own graphics card

In this video I show you how to run LLMs from Hugging Face using cloud infrastructure from Paperspace so you don't need to own your own GPU.
Note: This isn't in the video, but on the Model tab of the text-generation-webui interface, make sure the model loader is set to llama.cpp and n-gpu-layers is set to 256 so that the model is executed on the GPU, not the CPU.
Links mentioned in the video:
- Paperspace: www.paperspace...
- oobabooga's text-generation-webui repository on GitHub: github.com/oob...
- Example Hugging Face model: huggingface.co...