Give GGUF the same filename as project name

[sebaxakerhtc]

Why?
Default filename is "unsloth" - and that's nice, but when you have multiple models on HF and try to download them to OpenWebUI - they all have the same name like "unsloth.8.0.gguf".

Tested in colab with local and HF saving of gguf.

Local save

INFO:hf-to-gguf:Set model quantization version
INFO:gguf.gguf_writer:Writing the following files:
INFO:gguf.gguf_writer:/content/test-llama-3.2-1B/test-llama-3.2-1B.Q8_0.gguf: n_tensors = 147, total_size = 1.3G
Writing: 100%|██████████| 1.31G/1.31G [00:25<00:00, 51.9Mbyte/s]
INFO:hf-to-gguf:Model successfully exported to /content/test-llama-3.2-1B/test-llama-3.2-1B.Q8_0.gguf

HF save

Unsloth: Conversion completed! Output location: /content/sebaxakerhtc/test-llama-3.2-1B-GGUF/test-llama-3.2-1B-GGUF.Q2_K.gguf
Unsloth: Saved Ollama Modelfile to sebaxakerhtc/test-llama-3.2-1B-GGUF/Modelfile
Unsloth: Uploading GGUF to Huggingface Hub...
100%
 1/1 [00:05<00:00,  5.84s/it]
test-llama-3.2-1B-GGUF.Q2_K.gguf: 
 592M/? [00:05<00:00, 371MB/s]
Saved GGUF to https://huggingface.co/sebaxakerhtc/test-llama-3.2-1B-GGUF

[danielhanchen]

Oh agreed attaching the original name is better