Run #LLaMA 7B int4 model on #RISCV #LicheeRV (D1, [email protected])! Hack llama.cpp code to support running 4GB model with 0.6GB memory, no swap ...
確定! 回上一頁