Llama.ccp

I don't know about the Huawei hardware, but you can also try with some RISC-V chips. Recently I have been testing the SpacemiT K3, and they have a modified version of llama.cpp, to run on the dedicated RISC-V "AI" cores (1024-bit vectors). [https://github.com/ggml-org/llama.cpp/blob/master/docs/build-riscv64-spacemit.md](https://github.com/ggml-org/llama.cpp/blob/master/docs/build-riscv64-spacemit.md) [https://www.reddit.com/r/spacemit\_riscv/comments/1tcrx4o/collection\_of\_k3\_ecosystem\_hardware\_partner\_store/](https://www.reddit.com/r/spacemit_riscv/comments/1tcrx4o/collection_of_k3_ecosystem_hardware_partner_store/) It's not bad for a chip that has the performance similar to a Raspberry Pi 5. [https://youtu.be/3ziCQHnUqlE](https://youtu.be/3ziCQHnUqlE)