Question on quantization
#1
by Bercraft - opened
Hi what tool you used to quantize the model to gguf, since whisper.cpp only supports .bin fles. I am trying to quantize it but i cannot find a way. Could you tell me what tools you used? Thank you
Hi!
Whisper.cpp supports both the bin format and the old-style GGUF/GGML, which has been abandoned in Llama.cpp. In order to use it, you simply need to rename your resulting bin file to GGUF. And that's it
FerrisMind changed discussion status to closed