Request for support for Llama 2

[kennethleungty]

GGML version from TheBloke is coming soon: https://huggingface.co/TheBloke/Llama-2-7B-GGML

Thanks!

[marella]

It is working without any changes. Model architecture might be same as LLaMA 1:

from ctransformers import AutoModelForCausalLM

llm = AutoModelForCausalLM.from_pretrained("TheBloke/Llama-2-7B-GGML", model_type="llama")

print(llm("AI is going to"))

[kennethleungty]

Thanks!