While everyone's talking about Apple's Containerization framework announcement at WWDC, just a few days ago #krunkit quietly hit a major milestone: GPU passthrough in VMs, with up to 80% native LLM performance.

#krunkit and #podman are still the best hypervisor and container combination on macOS. #libkrun is also a great option for AI microVMs. We are thankful to be able to take advantage of these features in #RamaLama #AI

https://github.com/containers/ramalama

But I'm still looking forward to where Apple's variants go.

https://developers.redhat.com/articles/2025/06/05/how-we-improved-ai-inference-macos-podman-containers