Spaces:
Running
on
A10G
Issue converting FLUX.1-dev model to GGUF format
I am writing to report an issue I encountered while attempting to convert the FLUX.1-dev language model to the GGUF format using the tools provided in the ggml-org/gguf-my-repo repository.
I followed the instructions and used both the official FLUX.1-dev model and my own custom version, but in both cases, I received the following error: “FileNotFoundError: [Errno 2] No such file or directory: ‘FLUX.1-dev/config.json’”. It appears that the “config.json” file, which is crucial for the conversion process, is missing from the pretrained models.
I have researched this issue and found that other users have reported similar problems. I would appreciate it if you could investigate this matter and provide any solutions or additional guidance.
Thank you for your time and attention.
With custom fine tuned model you'll get "adapter_config.json" and adater_model, But you can save the configuration file during fine tuning like "config.save_pretrained('./path to model save')". This way you'll get config.json file as well in the repository.
I didn’t train the models. They are simply merges therefore don’t have config.json
any updates