https://huggingface.co/databricks/dolly-v2-12b

converted to mlc_llm format using https://github.com/mlc-ai/mlc-llm

command

python build.py --model dolly-v2-12b --target vulkan --quantization-mode int4 --quantization-sym --max-seq-len 2048

not fully tested because i got "CUDA: out of memory"