Request huggingface.co/neulab/Pangea-7B
#1474
by
sillykiwi
- opened
Please consider making GGUF version(s) of https://huggingface.co/neulab/Pangea-7B (BF16) or https://huggingface.co/neulab/Pangea-7B-hf (F16).
Thanks for all the work you do making models available for people!
I tried them booth but they unfortunately booth failed. Pangea-7B failed due to having an architecture not supported by llama.cpp. Pangea-7B-hf was more promising but still failed due to containing tensors that can't be mapped when converting the model into a source GGUF.