Files uploaded to minigpt4-7b-ggml and minigpt4-13b-ggml HF repositories are the same

Maknee / minigpt4.cpp

Port of MiniGPT4 in C++ (4bit, 5bit, 6bit, 8bit, 16bit CPU inference with GGML)

MIT License

549 stars 26 forks source link

Closed felladrin closed 12 months ago

felladrin commented 1 year ago

I noticed that the files uploaded to maknee/minigpt4-7b-ggml are actually the same that were uploaded to maknee/minigpt4-13b-ggml.

Also confirmed by running the model. It identifies minigpt4-7b-ggml with Model type: Vicuna13B.

Maknee commented 12 months ago

Wow, I didn't even realize this... Thanks for the catch! Probably pointed the 7B to the remote 13B folder when uploading...

I will convert 7B and upload the results. Will update when that is done.

Maknee commented 12 months ago

Updated. It should show up as 7B model now.