A guide for downloading base models, merging them with LoRA adapters, and deploying the result using Fireworks.
firectl
installed--quiet
: Suppress progress bar-h, --help
: Display help informationfirectl list models
and are denoted with the type HF_PEFT_ADDON
. Download a LoRA adapter using the same command as downloading a model.
adapter_id
with your desired LoRA adapteroutput_path
is a valid directory pathmerge_model.py
) with the following code:
base_model_path
to ./base_model/hf
because required files such as config.json
might be within the hf
directory.tokenizer_config.json
tokenizer.json
special_tokens_map.json
device_map="cpu"
firectl