Since NVIDIA owns huggingface now and huggingface has the excellent Candle [1] crate for inference on Rust, this seems like a good step towards nice native Rust kernels.[1] <a href="https://github.com/huggingface/candle" rel="nofollow">https://github.com/huggingface/candle
noob here - what's the benefit of this? Will using Rust lead to more optimal LLMs or code or both?
I view it more of supporting an expanding use case. If rust gets popular then you'll want to support it.
dllu · · focus · HN ↗
[1] <a href="https://github.com/huggingface/candle" rel="nofollow">https://github.com/huggingface/candle
instagraham · · focus · HN ↗
jvanderbot · · focus · HN ↗