How to convert LlavaForConditionalGeneration model with convert-hf-to-gguf.py

[forestsource]

python .\convert-hf-to-gguf.py --outtype bf16 "A:\llama.cpp\models\llava-calm2-siglip"
INFO:hf-to-gguf:Loading model: llava-calm2-siglip
ERROR:hf-to-gguf:Model LlavaForConditionalGeneration is not supported

[1171000410]

I encountered the same issue, have you resolved it?

[forestsource]

No yet.

[weilu4606]

the vision tower only support vit

[rtrad89]

I encountered this with a different model, Pixtral-12B

./convert_hf_to_gguf.py ./models/Pixtral12B
INFO:hf-to-gguf:Loading model: Pixtral12B
ERROR:hf-to-gguf:Model LlavaForConditionalGeneration is not supported

It looks like we have to wait see this