gregordinary/ggml-rocket

Llama.cpp with the Rockchip NPU integration as a GGML backend.

Fully open-source, high-performance LLM inference engine for the RK3588 NPU

Ollama alternative for Rockchip NPU: An efficient solution for running AI and Deep learning models on Rockchip devices with optimized NPU support ( rkllm )

Run RKLLM only in one docker command. Already converted models here.

Userspace driver, matmul, and on-NPU op library for Rockchip NPUs (RK3588 & Subset on RK3576) via the mainline rocket DRM-accel driver

Because RKNPU only knows 4D

LLM inference in C, compiled per model and per chip. Up to 14x faster.

Inference RWKV with multiple supported backends.