add "dim_model_base" to config.json to fix llama.cpp support

update README.md

update README.md

update README.md

update README.md

update README.md

update README.md

update README.md

update README.md

update config

add MiniCPM4-8B model

update README.md

initial commit

initial commit