search

modal-labs / llm-finetuning

Guide for fine-tuning Llama/Mistral/CodeLlama models and more
MIT License
538 stars 84 forks source link

how to download the finetuned model of the moda networkfilesystem to local? #22

Closed Qclover closed 8 months ago

gongy commented 11 months ago

This repo uses modal.Volume which is separate from NFS as specified in the title.

For large models, the recommended action is to upload it to S3/GCS if you want to move the weights out of Modal.

To download it, you can set up a Python http.server with modal.Tunnel with the volume attached and download the files through there.

mwaskom commented 8 months ago

Hope that solved your issue!