Max GPU: A new GenAI native serving stac(modular.com)
modular.com
Max GPU: A new GenAI native serving stac
https://www.modular.com/blog/introducing-max-24-6-a-gpu-native-generative-ai-platform
5 comments
Awesome release! It has been nice watching Modular work on enabling generic GPU programming. The level of work involved is really mind-boggling.
Has anyone tried Mojo and their other products? The language looked amazing but I remember many people being concerned about their things being closed-source. Taking a look now, it seems like it is mostly open sourced now?
Modular has an active community on Discord with a community showcase. They also have a new forum. Those are probably the best places to get a sense for what's being built with Mojo currently.
I believe Modular committed to open sourcing more components of Mojo at their last community meeting.
I believe Modular committed to open sourcing more components of Mojo at their last community meeting.
Sell me on this: if we already have a vLLM-based infrastructure that is working for us, what does this do or make easier that justifies the switching cost?
-Chris