Laya runs fully offline on a Mac M4, hitting 45 decisions per second
Laya on Mac M4 CoreML Offline
A GitHub gist from fordnox shows how to run Laya, an open-source decision model, entirely offline on a Mac M4 using CoreML and the Apple Neural Engine. The setup relies on the laya-coreml package and a multilingual CoreML model, installed with uv and downloaded via hf. The demo, a Snake game driven by the model, reaches roughly 45 decisions per second with no network connection.
Author testing https://github.com/mizorewww/laya-coreml
- imranq
Based on my admittedly limited research, it seems like you should use Laya for much more deterministic tasks where you have some training data. It won't be as good as Jev for zero shot cases.
- PaulRobinson
Local LLMs are the future, and one of the reasons I think the data centre furore is just going to end in a market crash.
LLMs that can reliably be used for control problems are the future, and I think classic/deep RL has generally been overlooked for years for a whole host of problems by wider industry because it felt inaccessible. The first thing I thought of when I saw Jev (and then Laya), was "this might move the needle in a really, really interesting way".
Local LLMs that can reliably be used for control problems smash through a lot of barriers I'm interested in, and this intrigues me a lot. Guess I'm about to become a big Laya fan if it can run on this kind of hardware to this performance.
- speedping
So cool. I've fired up pumas (energy monitor) and it seems to run almost fully on the neural engine and not the GPU so it plays really nicely with CoreML
- altano
How much memory does this use of the test machine's (M3 Max) 128 GB unified memory?
- tentacleuno
This looks like a local AI model playing Snake -- is that correct? The article offers no explanation.