#K3s

Ollama Crawling on K3s? A Practical Guide to GPU Passthrough Configuration

Ollama slow on K3s? The root cause is containers can't see the NVIDIA GPU, falling back to CPU. This guide breaks down the reasons and provides a complete GPU passthrough configuration: NVIDIA Container Toolkit, containerd template, Device Plugin deployment, and Ollama resource declaration, with verification and troubleshooting. Speed goes from 1 token/s to 60-90 tokens/s.