Help! Can't download models

Help! in LM Studio, I get that “Checksum failed. File corrupted” after the “Validating file integrity” step.

So then I thought, let’s manually download the models I want to try but I just realize it’s not that simple for the beginner that I am. On all the “model cards” that I clicked, no link to download the model, which I believe to be a .ckpt file. And when I check in the “Files and versions”, again, no way to just download 1 file, you need to individually download the different assets :

Could anyone please have a solution for the LM Studio problem OR could link me a step by step guide to how to download / compile / organize the files so that the AI is usable?


I assume you refer to this model: lmstudio-community/Meta-Llama-3-8B-Instruct-GGUF · Hugging Face.

LM Studio works with .gguf files (GGUF is a format to store large language models in an efficient, quantized way). Typically there are various ways of quantizing a model, e.g. Q8 refers to 8-bit quantization (storing each parameter using 8 bits or a single byte), Q6 refers to 6-bit quantization, and so on. Typically the fewer bits you use, the less performant the model will be (but also the more efficient it is to store on your computer).

Hence if you go to the “files and versions” tab of the model above, you can download one of the files, you don’t need to download them all.

Alright, thanks for the answer! I really was about to give up. Will try and update