This question hit me when I was downloading Galactica-1.3b model from https://huggingface.co/facebook/galactica-1.3b/. The only way I have found so far is to go to the model’s HuggingFace page and then click on Files and Versions. For example for Galactica https://huggingface.co/facebook/galactica-1.3b/tree/main
You should then see something like the following:
This will show the files in the model. Check pytorch_model.bin and on the right, it should show you the size of the model as pointed by the red arrow in the screenshot. In this case it shows you that the size is 2.63 GB.
Which also means that you need to run git lfs install before downloading the model before git clone https://huggingface.co/facebook/galactica-1.3b/
Of course, this only applies to open source models. If you know any other way to find the size of a LLM let me know in the comments. Thanks!
