But in the latest version of mlc-llm, there is no quantization option related to int8 quantization.
BTW, What is the difference between FT and Group quantization? It would be help a lot if there are docs about how this quantizaiton do.