How to convert fine-tuned model to GGUF ?

[heyili]

Hey guys,
I am trying to convert a fine-tuned model from huggingface jzdesign/falcon-finetune-midjourney-falcon .GGUF format. The fine-tuned one only contains a .json file that points to the base model and a .bin file
which contains all the weights.

my folder structure:
\fine_tune
-adapter_model.bin
-adapter_model.json

When I try to run

!python llama.cpp/convert.py fine_tune/ \
  --outfile fine_tune_convert \
  --outtype q8_0

I get

/content/llama.cpp/gguf-py
Traceback (most recent call last):
  File "/content/llama.cpp/convert.py", line 1658, in <module>
    main(sys.argv[1:])  # Exclude the first element (script name) from sys.argv
  File "/content/llama.cpp/convert.py", line 1577, in main
    model_plus = load_some_model(args.model)
  File "/content/llama.cpp/convert.py", line 1343, in load_some_model
    raise Exception(f"Can't find model in directory {path}")
Exception: Can't find model in directory fine_tune

So far I followed the tutorial from [https://github.com//discussions/2948]

In the end, what I would like to have, is a file like falcon-finetune-midjourney-falcon.gguf

Does someone know how to solve this problem ?

Thank you!

[asorteberg]

I am also seeking clarification on this process. Additionally, how can we use safetensor adapter files rather then bin files to reach the same conclusion.