Request of NeMo-Megatron-Launcher in other language for local inferencing
bm777 opened this issue · 0 comments
bm777 commented
It is a great framework, it performs very well on the cloud, would you also consider having this framework in Rust, I'm building a desktop application and the hybrid backend in Python and Rust, my combination has proven to be memory-safe. I have a monitor, inferencing on the client desktop app with at least 8GO of memory (RAM), which is always holding 1.78Go around 2Go.
Sorry for creating an issue, this repo doesn't have a Discussion section for this kind of question.