[Question] Would "Code Llama" be supported out of the box?

[akawrykow]

Meta just announced Code Llama which is a specialized model for code generation and discussion around code.

Code Llama is a code-specialized version of Llama 2 that was created by further training Llama 2 on its code-specific datasets, sampling more data from that same dataset for longer.

Since it is just a fine-tuned version of LLama 2, I'm guessing it should work out of the box with llama.cpp.

Can anyone confirm? If it works, should we add some examples/update docs/etc? 😄

[akawrykow]

Wrong repo, sorry will move this over to llama.cpp