rknpu

What problem does it solve?

This Skill removes the guesswork from deploying and operating the RK1 NPU inference pool, helping you avoid broken mounts, invalid model files, and unstable routing.

Core Features & Use Cases

• RK1 NPU device mapping for the correct renderD129 target and privileged /dev/dri access.
• Modelfile validation for rkllama using FROM and HUGGINGFACE_PATH in the expected key-value format.
• Pool operations and checks for four-node inference deployments, LiteLLM least_busy routing, and context-limit-safe configuration.
• Use case: bring up or troubleshoot a TuringPi 2 RK1 inference stack and confirm each node serves requests correctly.

Quick Start

Ask the assistant to deploy or verify the RK1 NPU pool for your cluster, checking device mapping, Modelfile format, context limits, and LiteLLM routing end to end.