GGML format please!
#8
by
dnhkng
- opened
Could we get this model in GGML, as for V2?
Or at least the conversion script :)
Thanks!
You can find some GGUF quants here: https://huggingface.co/FL33TW00D-HF/distil-whisper-large-v3/tree/main
GGML can be created using tensor-tools in candle whilst the script isn't integrated: https://github.com/huggingface/candle/tree/main/tensor-tools
GGML weights are available at: https://huggingface.co/distil-whisper/distil-large-v3-ggml
See instructions on README for running in Whisper.cpp: https://huggingface.co/distil-whisper/distil-large-v3#whispercpp