Can this model not recognize images?
#1
by
Wabifocus
- opened
Can this model not recognize images?
I don't think so. I use scripts to convert the models and it attempts to create a mmproj ( need edfor vision) for all models. In the case of this model that failed. either the model does not support this type of image processing or llama.cpp has not yet added support for it. Also checking the config files on the original model repo I see so image type tokens . If you find out otherwise please let me know and I will keep an eye out for llama.cpp support.