How do I convert models from q4_2?

[skidd-level-100]

In light of the current readme stating:
warning TEMPORARY NOTICE ABOUT UPCOMING BREAKING CHANGE warning

The quantization formats will soon be updated: #1305

All ggml model files using the old format will not work with the latest llama.cpp code after that change is merged
:
I would like to have a few command examples for when this change hits to convert my alpaca-enhanced model (https://huggingface.co/Pi3141/alpaca-7b-native-enhanced) to whatever the new format will be, also any other conversion commands
are great as I have a couple more models I like (Wizard-7B-uncesored, and more).

#1288
is empty ATM

[Green-Sky]

you could get the originals https://huggingface.co/8bit-coder/alpaca-7b-nativeEnhanced/ and convert them yourself. (that is all Pi3141 did in this case)

or just wait for ppl to roll out the new files.

[Green-Sky]

however, this model is relatively old. use something that is newer, like vicuna or wizardml etc..

[skidd-level-100]

you could get the originals https://huggingface.co/8bit-coder/alpaca-7b-nativeEnhanced/ and convert them yourself. (that is all Pi3141 did in this case)

or just wait for ppl to roll out the new files.

thanks for this but could you give me some command syntax --help does not help me and I cant seem to find enough online

[Green-Sky]

this section in the readme should get you going :)
https://github.com/ggerganov/llama.cpp#prepare-data--run