Does this support metal?

[bbecausereasonss]

LammaCPP introduced metal support, for M1/M2 blazing fast inference for GGML's.

[zmre]

Not the author, but this project relies on llm, which says this: "At present, inference is only on the CPU, but we hope to support GPU inference in the future through alternate backends." And llm depends on ggml, which has this open issue: metal support. So I'm guessing that means that metal isn't available in secondbrain... yet, at least..

[nightscape]

llm does support GPU inference on Metal. I'm using it frequently 😃 The docs might be outdated.