details of conversion process?

#1
by SQCU - opened

hey super curious: how did you convert the provided models to the gguf format?
the documentation in the llamacpp repository for /examples/llava are surprisingly opaque and out of date, making it very difficult to reproduce model training and quantization workflows seen in public.

Owner

Hi, I just quantized from the f16 gguf of xtuner/llava-llama-3-8b-v1_1-gguf.
For up-to-date llamacpp always check the discussions, PR, and issues.

MoMonir changed discussion status to closed

Sign up or log in to comment