nvidia/Hymba-1.5B-Instruct

#557
by SuperNintendoChalmers - opened

This small model is near the top of Big Hard Bench for edge-sized models, seems interesting enough to create GGUFs:

https://huggingface.co./nvidia/Hymba-1.5B-Instruct

Indeed, but somebody will have to write support for it in llama.cpp first :(

mradermacher changed discussion status to closed

Sign up or log in to comment