When I switch the model to the HF default cache path, the model list and the model loading page both display in a jumbled manner. I suspect it is due to the different ways the two organize their model structures, such as HF using -- (e.g., ~/.cache/huggingface/hub/models--meta-llama--Llama-3.2-1B-Instruct), while LM Studio uses directories / (e.g., ~/.cache/lm-studio/models/mlx-community/Llama-3.2-3B-Instruct-4bit), the method for compatibility with HF should not be too complex.
When I switch the model to the HF default cache path, the model list and the model loading page both display in a jumbled manner. I suspect it is due to the different ways the two organize their model structures, such as HF using
--(e.g.,~/.cache/huggingface/hub/models--meta-llama--Llama-3.2-1B-Instruct), while LM Studio uses directories/(e.g.,~/.cache/lm-studio/models/mlx-community/Llama-3.2-3B-Instruct-4bit), the method for compatibility with HF should not be too complex.